COMPETITION TRIBUNAL
REPUBLIC OF SOUTH AFRICA
Case Number: 21/LM/Apr01
In the merger between
NESTLE (SA) (PTY) LIMITED
and
PETS PRODUCTS (PTY) LIMITED
and
HEINZ SOUTH AFRICA (PTY) LIMITED
and
TIGER FOODS LIMITED
(“the parties”)
Reasons for the Competition Tribunal’s Decision
APPROVAL
1. On 31 May 2001 we approved the merger between the parties subject to certain
conditions in terms of section 16(2)(b) of the Competition Act 89 of 1998. We
deal with the reasons for our decision below. To give full effect to the conditions
attached to the merger approval we felt that it was necessary to keep some of them
confidential and known only to the parties and the Commission and ordered
accordingly. Since the reasons for our decision constitute a public document there
will be no reference in it to the confidential conditions. The nonconfidential
version of our order is annexed hereto.
THE MERGER TRANSACTION
2. Nestle (SA) (Pty) Limited (Nestle), a wholly owned subsidiary of the Swissbased
Nestle S.A company, is acquiring all the issued share capital in Pets Products
1
(Pty) Limited (Pets Products) together with the shareholders loan accounts. Heinz
South Africa (Pty) Limited (Heinz SA) and Tiger Foods Limited (Tiger Foods),
both companies incorporated in South Africa, hold 51,1% and 49,9% of the issued
shares in Pets Products, respectively. Accordingly Nestle SA is acquiring the
shares from Heinz SA and Tiger Foods.
3. Pets Products manufactures and sells wet pet food. It also sells dry pet food
manufactured for it by third party contractors. The majority of its products are
sold through the retail sector under the brands Butch, Husky, Dogmor, Catmor,
Pamper, Catz d Lite and Happy Cat. Its manufactures its wet food products at its
factory in Cape Town.
4. Through a wholly owned subsidiary, Friskies Petcare (Pty) Limited, Nestle is
involved in the manufacture and sale of dry pet food and imports and sells wet pet
food in South Africa. Nestle sells pet food to the retail sector under the Friskies,
Epol, Felix and Fancy Feast brands and under the Epol, Olympic and Expert
brands to the nonretail sector. 1 Its manufactures its dry food products at its
factory in Pretoria.
BACKGROUND TO THE PET FOOD INDUSTRY
5. The merging firms are both involved in the manufacture and distribution of pet
foods specifically for cats and dogs. The total retail value of this market is
estimated to be R 1,4 billion. The products in this market are differentiated first as
between cat and dog and secondly between dry and wet. Maize is the major
component in dry products and is usually sourced locally. Dry pet food constitutes
72% of the sales in the pet food market. Wet products are essentially based on
meat offal, or fish, but contain a cereal component. Over 60% of the wet food raw
material is imported or contains imported components. Wet food is typically sold
material is imported or contains imported components. Wet food is typically sold
in cans. Wet products constitute about 28% of sales (in value) in the pet food
market. Dog food constitutes 93 % of the market in volume (as opposed to value).
Imports account for approximately 7 % of pet food sales.
6. Industrial pet food is sold through both retail and nonretail channels with the
retail sector comprising 69.63 % of the total market.
THE COMMISSION’S RECOMMENDATION
7. The Commission recommended that this merger be approved unconditionally.
The Parties’ Submissions
1 The nonretail sector refers to products sold through outlets such as veterinarians, kennels, breeders,
pet shops, nurseries and state or industrial tenders.
2
8. The parties defined the relevant product market as comprising the sale of all pet
food in South Africa. They submitted that dry and wet pet food is interchangeable
and therefore belongs in the same market. They also claimed that the majority of
pets in South Africa are still fed homecooked food or table scraps and as a result,
it is premature to segment the market at this stage. The market for pet food is still
growing and the focus of the parties at the moment is persuading more pet owners
to switch to industrial pet food, as opposed to table scraps, without distinguishing
between dry and wet pet food.
9. If the market comprises all industrial pet food as the parties contend, Pets Products
would have a market share of 18% and Nestle a share of 10,5%. Accordingly the
merger would result in a market share of 28,5% for Nestle. The next largest
competitor Masterfoods has 23,1% of the market and thereafter the fourth largest
firm is Nola who have 6,7%. The parties submitted that the merger was not likely
to result in competition concerns because even post merger their combined market
share was relatively low. Furthermore, they argued that the market had very low
barriers to entry, there was a strong degree of countervailing power possessed by
their customers in the retail sector and they also faced competition from imported
products. They pointed out that there are a large number of small firms in the
market, albeit not necessarily national players, who constitute a competitive threat
to the larger firms. The parties submitted that all these factors diminish any
likelihood of the merger lessening competition.
The Commission’s Analysis
10. The Commission found that there is a clear distinction between:
10.1.1.cat and dog food;
10.1.2.dry and wet pet food;
10.1.3.dry pet food sold through the retail channel and dry food sold through
nonretail channel; and
nonretail channel; and
10.1.4.wet pet food sold through the grocery channel and wet pet food sold
through the nonretail channel.
11. Based on the above findings the Commission segmented the relevant product
market into eight submarkets consisting of the following:
11.1.1.dry dog food sold through the retail channel;
11.1.2.dry dog food sold through the nonretail channel;
11.1.3.wet dog food sold through the retail channel;
11.1.4.wet dog food sold through the nonretail channel;
11.1.5.dry cat food sold through the retail channel;
11.1.6.dry cat food sold through the nonretail channel;
11.1.7.wet cat food sold through the retail channel; and
11.1.8.wet cat food sold through the nonretail channel.
12. Table 1 below shows each participants’ market share in respect of sales value in
each of the relevant submarkets as identified by the Commission:
3
WET CAT
FOOD
DRY CAT
FOOD
WET DOG
FOOD
DRY DOG FOOD
Name of
party
Retail Non
retail
Retail Non
retail
Retail Non
retail
Retail Non
retail
Nestle 2,4% 33,1% 3,8% 3,5% 13,5% 7,4%
Pets Foods
39,5% 13,6% 66,5% 8,6% 2,5%
Master
Foods 27,4% 35,6% 47,5% 27,2% 8,8% 64,1% 18,3% 26,8%
Private
Label
6,1%
14,1% 6,9%
Promeal 4,2% 7,8%
Federal
Marine 20%
Hills 37,5% 22,1% 28,9% 16,8%
Iams 26,9% 35% 3,5% 17,8%
Vets
Choice 13,4%
Royal
Canin 10,3% 4,5%
Nola 1.1% 0,1% 13% 6,6%
Senwes 2,8%
Others 0,4% 4,7% 1,5% 2,8% 36,9% 4,2%
TOTAL 100 100 100 100 100 100 100 100
Table 1.
13. The shares of the merged entity in the various submarkets will therefore be as
follows:
WET CAT
FOOD
DRY CAT
FOOD
WET DOG
FOOD
DRY DOG
FOOD
Retail Non
retail
Retail Non
retail
Retail Non
retail
Retail Non
retail
Merged
Entity 41,9% 46,7% 3,8% 66,5% 3,5% 22,1% 9,9%
Table 2
4
14. The Commission found no likelihood of the lessening of competition in any of the
markets where products were sold through the nonretail channel. It was however
concerned that the merger would create high market shares for the merged entity
and increase concentration levels in the wet and dry cat food sold through the
retail channel market and in the market for dry dog food sold through the retail
channel.
15. Despite high concentrations in the above markets the Commission concluded that
the parties’ ability to exercise market power post merger would be constrained by
low barriers to entry, a strong degree of countervailing power possessed by the
retailers and the constraining effect on price of import competition.
16. In the dry pet food market the Commission found that entry was relatively
inexpensive and required no specialist knowledge. In the wet pet food market it
appears that although the cost of entry is higher than in the dry pet food product
market and more technical knowledge is required, the Commission was of the
opinion that this did not constitute an insurmountable barrier to entry. The only
regulatory requirement for entry into the market is registration in terms of the
Fertilizers, Farm Feeds, Agricultural Remedies and Stock Remedies Act (Act No.
36 of 1947) 2. As evidence of the absence of major barriers to entry the
Commission pointed to the existence of approximately 150 small manufacturers in
the dry dog food market and several bigger players who have entered both
markets over the last ten years. The Commission concluded that there are no
major barriers to entry in any of the markets.
17. A strong degree of countervailing power was found to exist in the retail sector.
The Commission argued that due to high concentration levels in the retail market
generally, suppliers are faced with huge retail customers and this limits their
negotiating power. Retailers interviewed by the Commission indicated that any
attempt to increase prices by any of the suppliers would result in them switching
either to other established brands or their own house brands. Some of the retailers
also stated they would consider importing products in the case of significant price
hikes whilst others thought this was not a realistic alternative. The Commission
was satisfied that the current import duty of 20% on pet food did not exclude
imports from competing with local products.
18. Finally the Commission accepted the parties’ argument that there was some
degree of competition between the parties’ products and home cooked food and
table scraps. It would appear that this argument was accepted by the Commission
more on the basis of homemade food constituting potential competition rather
than actual competition, i.e., were the parties to significantly increase the prices of
their products some of the pet owners who currently do not consider homecooked
food an alternative to whatever brand they are feeding their pets may change their
minds. This would explain why homecooked pet food did not form of the
2 Registration in terms of the Act is R700, 00 for each new product and a R350,00 annual renewal fee.
Approximately 167 firms are registered in terms of the Act.
5
relevant market as defined by the Commission but was considered only as one of
the factors that will reduce the likelihood of the high market shares and high
concentration levels in the market leading to a lessening of competition.
19. On the basis of the aforegoing the Commission recommended that we approve this
merger without conditions.
20. As mentioned above, we partially accepted the Commission’s recommendation
and approved the merger with conditions. The reasons for our decision appear
below.
THE RELEVANT MARKET
21. Both the merging parties and the Commission were in agreement that the relevant
geographic market was the whole of South Africa. We agree with this and need
not consider this aspect further. The debate between the Commission and the
parties concerns the appropriate product market. In determining a relevant product
market we have previously expressed our preference for the test set out by the US
Supreme Court in the Brown Shoe v United States 3. In that case the court held
that:
“The outer boundaries of a product market are determined by the
reasonable interchangeability of use [by consumers] or the cross
elasticity of demand between the product itself and substitutes for it.”
22. The court recognized that within the broad market as defined a number of “well
defined submarkets may exist which, in themselves, constitute product markets
for antitrust purposes.” According to the court factors which may indicate the
existence of a submarket include:
“industry or public recognition of the submarket as a separate
economic entity, the product’s peculiar characteristics and uses,
unique production facilities, distinct customers, distinct prices,
sensitivity to price changes, and specialized vendors.”
23. It is not necessary that all these factors exist and the US courts have found sub
23. It is not necessary that all these factors exist and the US courts have found sub
markets to exist even when only some of these factors are present. 4 In subsequent
decisions the US courts have held that the determination of the relevant product
market is:
“A matter of business reality… of how the market is perceived by
those who strive to profit in it.” 5
3 370 U.S. 294, (1962) at 325 We approved the Brown Shoe analysis in our decision in the merger
between JD Group Limited and Ellerines Holdings Limited CT:78/LM/Jul00 at page 123.
4 See FTC v Swedish Match (unreported US DDC Dec 14 2000 pg 14 case no Civ 001501)
5 FTC v Cardinal Health, Inc 12 F Supp 2d 34, (DDC 1998) at 46.
6
24. We have followed this approach in analyzing the market delineation in this case.
25. In the first place, the industry has created distinctive products for cats and dogs.
The products have different recipes; are marketed differently and have distinctive
brand names and get ups. All the major producers in the retail sector have separate
brands for cats and dogs. We have also had sight of the parties’ market research
material and business plans. These indicate clearly that the parties see this
distinction when they formulate their marketing strategies. There are also
functional distinctions and cat food is prepared in a manner to make it more
palatable6. Furthermore, on average cat food is more expensive than dog food.
Against this, other than the parties’ unsubstantiated assertions in these
proceedings, there is nothing in the evidence to suggest that the relevant market is
comprised of all pet food. We are satisfied that dog and cat products constitute
distinctive markets for competition purposes.
26. We are also satisfied that it is appropriate to distinguish between the markets for
wet and dry pet food. There are major differences in the production facilities
required for the manufacture of dry and wet pet food and the manufacturing
processes are very different. Furthermore, wet pet food is generally more
expensive than dry pet food, even when sold through the same channel. Products
also differ in volume, with dry dog food in particular being sold by some
manufacturers in 8 kg packs whilst wet food is normally sold in cans of much
lower volume. The Commission’s investigations also revealed that wet pet food is
perceived in the market as being something of a treat for pets and a supplement to
dry food rather than a substitute thereof. 7
dry food rather than a substitute thereof. 7
27. Lastly, we find that pet food sold through the retail channel belongs to a separate
market from pet food sold through the nonretail channel. Evidence before us
reveals a significant price difference between products sold through the retail and
nonretail channels. Sellers of products in the nonretail channel, for example,
veterinarians, are perceived by consumers as specialists and therefore authorities
in pet food. Consumers are therefore willing to pay higher prices for pet food
whose nutritional value is endorsed by these sellers compared to food sold through
the retail channel. 8 Significantly certain manufacturers who distribute to both
channels have different brands for each of them. Nestle markets its nonretail
product under the Olympic brand and Pets Products under the IVD Life Stages
brand.
28. We therefore agree with the Commission’s delineation of the market as set out.
6 According to Nestle “ For cat food palatability is the primary determinant. Cat food is on average
more expensive because the palatability demand is higher and for higher quality more expensive
materials are used.” (Letter to Tribunal dated 25 May 2001)
7 See comments of Dr Catton of the SA Veterinary Association interviewed by the Commission. Mr.
Marius Van der Vyfer also interviewed by the Commission stated that the SPCA does not feed with
wet pet food as it is too expensive and may cause diarrhea.
8 For instance Dr Catton from the Veterinary Association who was interviewed by the Commission
confirmed that products sold through the veterinary channel are more expensive because they are
scientifically engineered, have more nutrients and pets are fed less.
7
29. Indeed there may well be an argument for segmenting the market further between
price bands. Brands are targeted at three retail price bands Super Premium,
Premium, Mainstream and Economy.
Analysis of the market share concentrations
NonGrocery Channel
30. Regarding the nongrocery channel, the parties have no products in the wet cat
food segment of the market; there is no overlap in the dry cat and wet dog
segments of the market and both parties are minor players in the dry dog food
segment of the market. This merger is therefore not likely to cause any
competition concerns in the nongrocery channel market.
Retail Channel Dry/Wet Cat Food Market
31. We now calculate the HHI in the different submarkets based on the figures
supplied to us by the parties. 9 In the wet cat pet food market the current HHI is
2771.62, increasing by 190 to 2961.22 post merger. In the dry cat pet food market
the merger will result in an HHI increase of 900.32 from 3538.03 before the
merger to 4438.35 post merger.
32. The HHI figures for the wet and dry cat food retail markets both demonstrate an
increase in concentration sufficient to cause concern based on most conventional
analysis.10
Retail Channel – Wet Dog Food Market
33. The HHI in the wet dog food market remains unchanged at 4767.18. Since there is
no overlap in the wet dog food retail market this figure requires no further
comment.
Retail Channel – Dry Dog Food Market
9 Note that “Private Label” in table1 includes all supermarket house brands; as a result the HHI will
tend to be higher in markets where the house brands have significant market shares and we have taken
cognizance of this in our analysis. We have excluded the “other “ category from our analysis as this
represents companies holding less than 1% of the market.
represents companies holding less than 1% of the market.
10 According to the United States Department of Justice’s 1992 Horizontal Merger Guidelines if the
post merger HHI exceeds 1800 and the merger increases the HHI by 100 points or more the merger
will be presumed to create or enhance market power or facilitate its exercise. This presumption may be
rebutted by a showing that “nonstructural factors reveal that such an exercise of market power is
unlikely”.
8
34. The dry dog pet food market shows an increase of 232.20 from 815.55 to 1047.75
resulting from the merger. At first blush this HHI figure for the dry dog food retail
market appears innocuous. Yet a closer examination of price ranges within this
market and the sales figures in supermarket outlets based on AC Nielsen figures
presents another picture 11.
35. There are three submarkets for dry dog food in the retail channel economy,
medium and premium. The lowest priced product in the economy range is priced
at R2.30 per kg whilst at the premium end the highest priced product is R13.99 i.e.
six times as expensive. Price competition between these products is hardly likely,
given this differential. It is more likely that products compete within their
respective price category bands as these products are priced far more closely
within their bands.
36. Examining the dry dog food market in this way reveals that the market is more
concentrated in the medium and premium priced bands. In the medium price range
only four manufacturers are represented Nestle, Pets Products, Masterfoods and
Nola. In the premium price range we find only two manufacturers, Nestle and
Masterfoods.
37. The merger would mean that there would now be only three manufacturers
represented in the medium range whilst the premium range, with only two firms,
would face less likelihood of entry from an incumbent in the medium band.
38. We have also been provided with figures for supermarket retail outlets, which are
a submarket of the retail sector, but one that represents the bulk of sales through
the retail channel. These figures show on a four firm concentration ratio that the
market is more concentrated than the earlier HHI would suggest. On these figures
based on Oct 99 – Oct 2000 average sales figures by kilogram, the top four firms
have 62 % and the merging firms combined share of this is 31.5%. By rand value
have 62 % and the merging firms combined share of this is 31.5%. By rand value
these figures show an even greater degree of concentration with the top four firms
enjoying 73% of the market with the merging firms combined comprising 33.7%.
39. Using these figures to calculate an HHI for the supermarket segment of retail dry
dog food we find that the premerger HHI is 1423.31 and the post merger HHI is
1979.15, an increase of 555.84. These figures take one into the zone of
competitive concern.
40. The merged firm will own three of the six brand names that compete in the
medium and premium market. 12
41. Its is true that two major competitors to the merged firm will remain in the market,
Nola and Masterfoods both of whom have strong brands. Yet the pricing data
shows that Nola products are situated in the economy and low end of medium
range (they have no products in the premium range) whilst Masterfoods has no
11 The Nielsen survey only reads the major retail outlets and does not include figures for other retail
outlets such as convenience stores and garage forecourt stores.
12 Epol, Husky and Dogmor. Masterfoods has Pedigree and Pal and Nola has Bobtail.
9
products in the economy range and its prices range from high end of medium to
premium. Thus these firms will probably constitute separate rivals to the merged
firm at different levels of the pricing bands but do not seriously compete with one
another. This means that post merger one will have only two national firms in
effect in the respective pricing bands in the retail channel.
42. The A.C. Nielson data suggests that the market penetration of the house brands
and the other players is limited. None of the house brands holds in excess of 5
%13 of the market whilst the fifth largest firm, Sentraalwes, saw its market share
decline to 4%. The major producers, including Pets Products and Nestle,
manufacture the house brands. This relationship would suggest that they do not
constitute a significant competitive threat.
43. We conclude that the merger gives rise to concentration concerns about the dry
dog food market as well.
Anticompetitive effects
44. The mere fact that the markets are highly concentrated does not mean that a
merger is likely to be anticompetitive. We need to consider other factors as well.
The likelihood of the merged firm having market power and hence the potential to
raise prices post merger to a supracompetitive level is increased by the following
considerations:
44.1Prices of pet foods are relatively inelastic for certain consumers as once
owners have established that a pet has a certain preference they are reluctant
to change merely because of a small but significant price increase. This is
something the merging parties themselves have conceded in relation to cat
food although they are reluctant to make the same concession with dog food
falling back on the argument that owners would move to another brand. 14 Yet
it is difficult to see why on this basis the convenience considerations of the
dog owner and cat owner should be any different. Thus we find that price
dog owner and cat owner should be any different. Thus we find that price
sensitivity is not sufficient to inhibit a possible price increase.
44.2Pets Product and Nestle have been vigorous competitors in certain of the
markets identified. The pricing data shows that brands such as Dogmor and
Husky compete with those of Nestl é’s Epol and similarly Pamper and Catmor
compete with Friskies and Epol Cat. This also emerges from the market
survey analysis which taken cumulatively suggests that only three firms are
major competitors in three out of the four identified retail markets.
13 This figure is consistent with the parties estimate that house brands constitute 5,1% of the entire pet
food market.
14 In their memorandum to the Tribunal Nestle states “ Once a cat owner finds a product that the cat
likes he/she is loathe to change and thus price sensitivity is much lower. Brand switching is more likely
to happen because the cat has become bored with what is served or merely because the cat owner
wanted to try something different which the cat then prefers.” The memorandum goes on to state that
with dog food the issue differs and the only criteria is price. However later on they concede that the
reason the non – retail channel has become so popular and which offers products at much higher prices
is “an indication that for many people the price sensitivity is low.”
10
45. Our concerns that the merger will have anticompetitive effects can of course be
rebutted if there is evidence that notwithstanding the high levels of concentration
and anticompetitive features we have identified the markets remain contestable.
This is the Commission’s view because they argue there is countervailing power
in the retail market and there are low barriers to entry. We will now proceed to
examine these arguments.
Countervailing power
46. The merging parties claimed that they are faced with a large degree of
countervailing power where they sell their products through the retail channel.
The majority of the products sold through this channel are sold through the major
retail chain stores, like ShopriteCheckers, Hyperama, Spar and Pick ‘n Pay. These
retail outlets account for the overwhelming majority of sales in the retail channel.
47. In their submissions the parties claim that the retailers are very powerful and the
negotiation power of the manufacturers is limited. As a result, the manufacturers
are generally allowed to revise their prices once a year, usually at the end of
February, when the majority of the retailers had done their stocktaking for the
year. The parties claim that low margins and the high cost of doing business in the
grocery channel has driven most manufacturers, including Nestle, into finding
alternate channels of sale for their products. Manufacturers have to pay for shelf
space and advertisements, and grant logistic allowances and volume rebates to
retailers. The retailers also generally dictate temporary price promotions to the
manufacturers. The parties acknowledge that, strictly speaking, the manufacturer
has a choice whether or not to reach agreement with the retailer on the above.
They claim, however, that failure to do so generally results in delisting or
reduced orders and shelf space for the manufacturer concerned. From the parties’
submission, it appears that if a manufacturer cannot reach agreement with the
major retailers at the end of February in any year it faces virtual exclusion from
the market until the retailers do their next stocktaking at the of February of the
following year. The parties claim that the power possessed by the major retail
outlets means that they will be able to resist any price increases by the merged
entity. According to the parties, the retailers possess a degree of countervailing
power sufficient to restrain any anticompetitive pricing by the merged entity.
48. We do not accept this argument. While it may be true that the retailers are in a
powerful bargaining position with regard to listing, promotions, volume rebates
and the provision of shelf space, they have no incentive to resist price increases by
the manufacturers. Firstly a price increase by a manufacturer to all the retailers
will simply be passed on to the consumer, and therefore the profit margins of the
retailers will not be adversely affected. Secondly, some of the retailers have their
own house brands that compete with some of their suppliers’ products, a price
increase in any of the suppliers’ products means either that the retailers’ product
becomes more competitive or it allows the retailer an opportunity to increase
prices for its products as well.
11
49. Our view is that far from having a countervailing effect on the market,
supermarkets actually serve as a barrier to entry. From the above submission by
the parties it is clear that a new entrant wishing to mount a challenge for market
share against the established firms would need to get its products into the shelves
of the major retail chain stores. If large manufacturers with established brands are
finding it difficult to get their products onto the retail shelves what are the chances
of a new market entrant with an unknown brand succeeding in this market? If the
retailers’ position is very strong in the market against manufacturers of established
brands, their position is even stronger against a new entrant with an unknown
brand because of the significant role brands play in this market. Where retailers
can afford to delist established brands from their shops, a new unknown brand
stands very little chance of surviving in that market without a willingness to
sustain huge sunk costs in developing its brand, thereby eventually improving its
bargaining position against the retailers. As already indicated above the margins in
this market are very low, as a result most firms would not be willing to enter the
market if they have to sustain vast sunk costs to develop a brand.
50. It was also argued by the parties that a large degree of supply side substitution
exists in the market manufacturers, without changing their existing
manufacturing facilities, may switch between wet dog and cat food and between
dry cat and dog food. Any manufacturer who already has a brand in the either dry
or wet cat/dog food in the retail sector is quite capable of switching its supplies
between dog and cat food using the same brand to market both products.
Furthermore, the parties argued that if the prices of their products increased
significantly post merger some of the big international players with established
brands currently supplying only the nonretail sector, including Colgate Palmolive
and Proctor and Gamble, may start supplying the retail sector as well. While we
think that there is some merit to this argument we are not convinced that potential
supply side substitution is capable of offsetting the lessening of competition that is
in our opinion likely to result from this merger.
51. At the moment imports only account for 7% of the total pet food market indicating
insufficient penetration to constitute a serious competitive threat.
52. The likelihood of a new entrant establishing manufacturing operations in South
Africa is also remote unless it does so by acquisition. The merger itself is
evidence of this. Nestle has stated that it would build a new plant valued at a cost
of approximately R 80 million provided the merger was approved. Without the
additional throughput of Pets Products such a project would not be viable, as the
critical volumes required for such expenditure would not be met. According to
one of Nestl é’s representatives at the hearing the only way someone new would
enter is if they could expand in the African region. If Nestle, an incumbent, is
hesitant about expansion without a significant acquisition then a new entrant is
even less likely to do so.
53. Supply side substitution between wet and dry food is also not an option as
according to the parties, this is not possible technically.
12
Barriers to entry
54. As appears in paragraph 14 above the Commission found that there are no major
barriers to entry into the general pet food market. The Commission cited the legal
barrier of registration in terms of Act 36 of 1947 and startup costs as the only
hurdles to be overcome by an entrant into the market. We agree with the
Commission that registration and the costs of setting up a factory do not constitute
insurmountable barriers to entry into the market. Our opinion, however, is that the
costs of setting up a factory and registration of products are not the only barriers
to entry in this market.
55. Firstly we have already found that the concentration of power in the hands of
major retail chain stores on the retail channel constitutes a major barrier to entry
into the market. Furthermore, in interviews conducted by the Commission with
retailers to get their views on this merger, some retailers stated that they
welcomed the merger because it would result in them stocking fewer brands than
is currently the case. This alone may not be conclusive proof that retailers will not
welcome a new brand into the market, but their stated preference for dealing with
fewer manufacturers indicates that a new entrant will probably find it difficult to
acquire shelf space in their shops.
56. Secondly it is our view that the role of brands in this market is a huge barrier to
entry. The parties themselves have stated in their submissions that:
“The most important fact in the relationship with retailers is the brand.”
57. The big pet food manufacturers such as the parties, Masterfoods and Nola spend
vast amounts of money on advertising to promote their brands. Retail sector
advertising is responsible for by far the largest share of the advertising
expenditure incurred by the parties. For example, Pets Products spent a total of
expenditure incurred by the parties. For example, Pets Products spent a total of
just over R9,5 million on advertising last year; the retail sector accounted for
about R8, 9 million of this expenditure. Nestle spent a total of just over R18,3
million last year on advertising of which R14 million was spent on advertising
products sold through the retail sector. At the hearing, in discussing the decline of
their firm’s performance, the Pets Products representatives complained that
Masterfoods had invested much more in marketing then they had been able to.
58. Presumably a new entrant wishing to establish a new brand in the market in
competition with the parties’ brands would have to spend at least similar amounts
on advertising to promote its brand. In a market with solid brands and very low
margins very few firms are likely to do so. This may explain why there are a large
number of smaller manufacturers who are generally unable to secure shelf space
in the major retail chains and whose market share is stagnant. None of them has
been able to challenge the major manufacturers 15.
15 The parties cite Masterfoods as an example of a manufacturer who is a recent entrant
(approximately five years ago) and whose market share has increased rapidly. The case of Masterfoods,
however, serves only to indicate that at the time that it entered the market conditions were different to
what they are now. At the time that Masterfoods entered the market, Nestle itself had not yet entered
13
59. It is our opinion that the significant role played by brands in this market is a major
barrier to entry.
Conclusion
60. We have in the previous sections identified competition concerns that may arise as
a result of the merger. These concerns are not sufficient to justify prohibiting the
merger as long as the markets remain contestable.
61. What emerges from the analysis is that brands play a strong role in the market,
particularly in the retail sector. This means that effective entry into the retail
sector requires either the purchase of an existing brand or the introduction of a
new brand, which requires considerable investment.
62. Given the fact that most of the strong brands are owned by a small number of
firms it is unlikely that they would sell unless they exited the market or one of the
four submarkets. They are unlikely to sell a brand in a market in which they
continue to compete. Thus entry through acquisition is not likely to lead to an
increase in the number of players in the market.
63. Although manufacture would ordinarily constitute a barrier to entry it is not an
insurmountable one. A firm entering the dry food market will find that there is
sufficient existing local capacity whilst importing has not constituted a major
barrier for those who have taken that option particularly those in the wet food
market. Thus the crucial issue is brands.
64. Entry by the introduction of a new brand is unlikely to occur unless the entrant is
already in another submarket or on the edge of the market e.g. a firm positioned
in the nonretail sector.
65. Since margins in the retail pet food industry are small, most firms would be
unlikely to enter if it meant vast sunk costs having to be incurred in developing a
brand. The firms likely to enter are either those that, like Nestle, wish to use a
South African operation to expand in the rest of the region or those who have an
South African operation to expand in the rest of the region or those who have an
international brand which could be used locally which would reduce the cost of ad
spend.
66. No obvious candidates emerge in either category. Proctor and Gamble and
Colgate Palmolive, although both giants in the world of consumer brands, have
chosen to enter the nongrocery brand. They have chosen to use their brands
through this channel and, with some exceptions, it appears that the industry’s
collective wisdom is that different brands are required for the different channels.
67. If, however, strong retail brands became available, the prospect for entry is
enhanced if the firms in the market were to use their market power unilaterally or
nor had the joint venture between Heinz and Tiger Foods in respect of Pets Products been formed as
both occurred four years ago.
14
collusively to increase price above a competitive level. For this reason we have
decided to order the merging firms to divest two of their brands as a condition for
the approval of the merger.
68. The purpose is to encourage entry into the market by lowering the most significant
barrier to entry in the retail channel – ownership of a viable brand.
69. The two brands have been identified after consultation with the merging firms.
We have been careful to identify brands that do not constitute the crown jewels of
the merger, for that would be unfair to the merging firms, but nevertheless could
be sold as viable brands to a third party. The two brands are Dogmor and Catmor
both presently owned by Pet Foods. Although Catmor is at the moment only in the
dry cat food market it appears that a number of manufacturers use the same brand
for both wet and dry products and this is something consumers are used to.
70. In the dog food market Dogmor is used as a brand in both dry and wet products.
The common “mor” suffix to the two names means that they are also associated in
the mind of the consumer as the products of the same manufacturer, so they are
useful to sell as a package to the same purchaser.
71. The conditions attached to the divestiture appear from our order and are self
explanatory. We have however retained confidentiality over the procedural
aspects of the disposal in order not to prejudice the merged firm in its efforts to
dispose of the brands at a fair value.
72. One important condition is that the purchaser must be a viable one and hence we
have required the divestiture to be conditional on the approval of the Commission.
A sale to a firm that does not have the expertise and resources to usefully exploit
the brands would make this remedy meaningless. The purpose of the remedy is
not to help create a competitor with production capability and market share, rather
not to help create a competitor with production capability and market share, rather
it is to facilitate entry into the market so it remains contestable. 16 This means
identifying a firm with the ability to seize the opportunity and not one that will
squander it.
_____________ 18 June 2001
N.M. Manoim Date
Concurring: S. Zilwa; M. Moerane
16 See “A Study of the Commission’s divestiture process” , Federal Trade Commission 1999, page 6.
Here the Commission explains how its licensing remedy differs in its object from its past policy as the
licencing remedy’s effectiveness depended on the resources and knowhow of the firm which was to be
the licencee.
15