COMPETITION TRIBUNAL
REPUBLIC OF SOUTH AFRICA
Case no.: 46/LM/May05
In the large merger between:
Tiger Brands Ltd
Ashton Canning Company (Pty) Ltd
Newco
and
Langeberg Foods International
Ashton Canning Company (Pty) Ltd
Reasons and Order – Nonconfidential version
________________________________________________________________
Introduction
1. We approved this transaction, which concerns the merger between two
fruit canning businesses, Langeberg Foods International and Ashton
Canning Company (Pty) Ltd on 30 September 2005, subject to conditions
that relate to employment and appears at the end of this decision. The
reasons for our decision are set out below.
Witnesses
2. The Tribunal called the following witnesses during the hearing:
1) Mr Bruce Henderson of Rhodes Food Group (Pty) Ltd
2) Mr Oelof Janse van Rensburg of Breede Valley Fruit Processors
(Pty) Ltd
3) Mr Wehan Victor of The Canning Fruit Growers Association
4) Mr Hannes Hofmeyer of Del Monte South Africa (Pty) Ltd
3. The merging parties’ witnesses were:
1) Mr Wynand du Plessis of Ashton Canning
2) Mr Rudolph Richards of the South African Fruit and Vegetable
Canners Association
3) Mr Athanasios Martalas of Langeberg Foods International
The merger transaction
4. Newco, a special purpose vehicle, will acquire the Langeberg Foods
International (“LFI”) canning business conducted by Tiger Food Brands
Ltd (“Tiger Brands”) and the fruit canning business conducted by Ashton
Canning Company (Pty) Ltd (“Ashton Canning”).
5. In consideration for this, Tiger Brands will acquire two thirds of the shares
in Newco. Ashton Canning will acquire the remaining one third, as well as
certain minority protections or veto rights, which confer upon it the right to
materially influence the policy of Newco. Tiger Brands and Ashton
Canning will therefore jointly control Newco after the merger.
The parties to the merger
6. Tiger Brands is a branded food and healthcare group that is involved in
the production of fast moving consumer goods and healthcare products.
Whilst no single shareholder controls Tiger Brands it controls 20
subsidiaries of which, for our purposes, Tiger Food Brands Limited (“TFB”)
is the most important. TFB in turn wholly owns Langeberg Foods
International (“LFI”), the canning business involved in this transaction.
7. Ashton Canning, which controls Ashton Canning Industrial Concentrates
(Pty) Ltd, has the following shareholders:
• Ashton Canning Investments (Pty) Ltd 61%
• JF Barnard (Mark) Trust Fund 16%
• JF Barnard (Leta) Trust Fund 8%
• JF Barnard (Jeanne) Trust Fund 8%
• Estate of late R Barnard 7%
2
8. Ashton Canning, like LFI, is involved in the production of canned
deciduous fruit and fruit puree for the local and export market. 1
Rationale for the proposed transaction
9. According to the merging parties the deciduous fruit canning industry is an
industry in dire straights due to the strengthening of the rand, which has
made their products less price competitive in foreign markets. Since 80%
of their product is exported, and only 20% is sold locally, they need to
consolidate in order to lower their production costs and hence to enable
them compete profitably on the international market. 2 They claim that in
doing so they are following international trends set by international
retailers and other food service businesses.
10. Whilst this is the stated rationale for the merger, whether it is the sole
rationale is something we examine later.
The relevant product market
11. The merging parties activities overlap in respect of the production of
canned deciduous fruit and fruit puree concentrates.
12. Canned fruit is produced in five different can sizes and consists of canned
apricots, peaches, pears, guava, apples and fruit cocktail which are
produced in a variance of styles (halves, slices, quarters and dices) and
packing mediums (sugar syrup, natural fruit juices and artificial
sweeteners). Puree concentrates are mainly produced in large containers
for reprocessing by manufacturers of fruit juices, dairy products and jams.
13. There are four deciduous fruit canners in South Africa. They are Ashton
Canning situated in Ashton, Cirio Del Monte in Tulbagh, RFF Foods in
Drakenstein and LFI in Ashton. The canners purchase fruit from farmers in
the vicinity of their respective plants, typically supplied in terms of a
contract between the canner and the individual farmer.
contract between the canner and the individual farmer.
14. The canners in turn sell their canned product to local and foreign retailers,
1 Both merging parties are also involved in the production of canned baked beans and tomato paste.
However, since their market shares are less than 1% we will ignore it for purposes of this transaction.
2 85% of the canned fruit is exported and 15 % is sold locally while 75% of the puree is exported and 25%
sold locally.
3
brand owners who distribute their own brands and importers or
distributors. In the case of puree the canned puree is sold to other
manufacturers.
15. LFI owns the KOO and Naturlite brands, which are sold locally. Its
international brands are Gold Reef and Silver Leaf. Unlike its competitors
Ashton Canning, produces canned fruit on behalf of third parties and does
not own any of its own brands. It produces primarily for M&L Distributors
(Pty) Ltd, which owns the following three local brands:
• Farmgirl which is sold in Pick ‘n Pay retail stores,
• Lesmel which is sold in Shoprite retail stores, and
• Mayfair
16. The remaining deciduous fruit canners in the market that own canned fruit
brands are Del Monte, trading under the Del Monte and Layco labels and
Rhodes, which owns the Rhodes, Tartan and Hazeldene brands. Both
Rhodes and Del Monte also produce canned fruit for the private labels of
the three local retail chains.
17. Fruit puree is produced for reprocessing by manufacturers of fruit juice,
dairy products and jams. The canners produce puree by pulping the fresh
fruit, concentrating it through evaporation and filling it into bags in
steeldrums. It is sold unbranded and there is no substitute available for
fruit puree.
18. A number of competitors, in addition to the merging parties, compete in
this product market namely RFF, Del Monte, Boland Pulp and Breede
Valley River Processors (“Breede Valley”). Breede Valley entered into this
market fairly recently, in 2004, and has since then captured 5% of the
market.
19. It was common cause between the Commission and the merging parties
that these products form discrete product markets and have no
competitively significant substitutes. Thus though fresh fruit may be
functionally substitutable for canned fruit it is not considered a relevant
functionally substitutable for canned fruit it is not considered a relevant
substitute for the purposes of defining the market because of its high rate
of perishability.
The relevant geographic market
4
20. In respect of both markets, the Competition Commission has defined the
geographic market as national whereas the parties argue that it is an
international market.
21. According to the merging parties 85% of their canned deciduous fruit
product and 75% of their puree are exported to countries in Asia, the EU,
the Middle East, the Far East and Canada and they, accordingly, compete
with companies located throughout the world. They also claim that the
local market is constrained by the threat of actual and potential imports
particularly within the current and now sustained strong rand environment,
and that imports effectively compete with locally produced products.
22. The Commission accepted that the players, all of whom are based in the
Western Cape, compete in both national and international markets within
each product market. However, it found that import penetration into South
Africa was limited in the canned fruit market and nonexistent in the puree
market and, therefore, defined the geographic market as a national
domestic market for both products. 3 As the concentration levels differ
materially depending on whether one defines the market as a domestic
national market or an international market it is necessary for us to decide
this issue.
23. In deciding whether a market should be considered international because
of the presence of imports we held in Trident Steel that:
“ The mere presence of imports does not necessarily indicate
conclusively that a market is international. Some writers and indeed
other competition authorities take the approach that a market may be a
national one punctuated with sporadic sources of supplies for
overseas. This however would not necessarily warrant delineation as
an international market.” 4
24. In deciding whether a market should be defined as national or
24. In deciding whether a market should be defined as national or
international the Office of Fair Trading in the United Kingdom, suggests, in
a research paper on market definition, that the following factors should be
taken into account: 5
• Parallel price evidence between the areas, which are claimed to be
3 We were told during the hearing that imports represent 1% of total consumer packed canned fruit. See
page 20 of the transcript dated 13 September 2005.
4 See Trident Steel (Pty) Ltd v Dorbyl Limited 89/LM/Oct00 at paragraph 27.
5 See Office of Fair Trading Research Paper on Market definition in UK competition policy , by National
Economic Research Associates (NERA), February 1992.
5
part of the same geographic market
• Trade flow data
• Trade barriers
• Exchange rate factors and international uncertainty
25. The research paper also notes that in considering these factors one
should also keep in mind that the international market carries a higher
uncertainty factor than expectations concerning domestic affairs and this
uncertainty justifies the use of a stronger burden of proof before accepting
that the relevant market is international. As these are useful indicators to
apply to the facts of this case we consider some of them below.
Parallel price evidence
26. The merging parties have provided us with the international and domestic
price data for each commodity for the period 2001 to 2005. During this
period the exchange rate of the rand, particularly against the dollar has
proved to be extraordinarily volatile. 6 In 2001 the average exchange rate
for the year was R8.06 to the dollar depreciating to R10.85 in 2002.
Thereafter the value of the rand against the dollar appreciated sharply to
R9.46 in 2003, R6.62 in 2004 and R6.00 for the first half of 2005. Perusal
of the price data for LFI and Ashton show that the relationship between
international and domestic prices was significantly altered by these
changes in the exchange rate. 7 The international and local prices did not
move in tandem, the local prices increased while the international prices
were decreasing. And when the local prices reached a point where it was
higher than the international prices we find Del Monte reentering the local
market.
27. As the rand depreciated during the period 2001 to 2002 LFI was able to
charge higher prices, in rand, in the international market than in the
domestic market. 8 After 2002, as the rand appreciated, the relationship
between domestic and international prices altered with LFI’s domestic
between domestic and international prices altered with LFI’s domestic
prices now in excess of international prices. Ashton increased its local
prices above its international prices much later, in 2005, when the
exchange rate was R6 to the dollar. 9 It is clear that LFI and Ashton are
6 These data can be found on page 188 of the record. See summary attached as Annexure B to this
decision. This document is not attached as it contains confidential information.
7 See page 761 and 762 of the record
8 This provides evidence that the pass through of the exchange rate in Rand was reasonably complete and
that LFI was operating as a price taker on international markets.
9 This might be because LFI’s KOO brand is the leading brand locally and much stronger than those
6
able to segment the international from the domestic market as rand prices
domestically fails to be affected by the strong appreciation of the rand.
Domestic consumers were insulated from these exchange rate changes
suggesting that the two markets could be differentially serviced by
canners. The data therefore support the hypothesis that there is a
domestic for those canned fruits sold nationally that is separate from either
an international market or a series of foreign markets such as Japan, EU
and North America.
Trade flow data
28. In order to define a market as international it is suggested in the NERA
Research Paper, referred to earlier, that significant levels of import
penetration, at least 15% of domestic sales, should be required before one
can regard import competition as capable of exerting decisive influence on
price levels.
29. In this case neither the canned deciduous fruit market, claiming imports of
1%, nor the fruit puree market, having none, have recorded any imports
worth considering. There is little probability that imports will constitute a
large part of our domestic market for a sustainable period given the
competitive advantage of local firms not just in location, but also in terms
of quality.
Exchange rate factors
30. The merging parties argue that local producers are constrained by the
threat of actual and potential imports, particularly within the current strong
rand against the dollar, therefore, the geographic market should be
regarded as global.
31. We are not convinced by this argument since imports are negligible even
with the current strong rand. In fact, there is no indication in the evidence
that parties are concerned about the threat of imports due to the
strengthening rand. It is mostly in relation to exports that this issue is
raised by the merging parties. 10 The key to this might be found with the
South African consumer. According to Dr D.P. Keetch of Goldhamer
Consulting, “c onsumers [in South Africa] have become wary of buying
“cheap” products because they are considered to be of an inferior quality
to the wellknown brands. ”11
canned by Ashton.
10 See record page 802.
11 See the report by Dr Keetch: The South African Canned Fruit and Vegetable Industry , page 524 of the
7
32. It is also worth mentioning that the South African Fruit & Vegetable
Canners Association in its report on “ Growth Opportunities In the
Deciduous Canned Fruit Industry ” distinguishes between an international
and a local market. 12
33. Furthermore this conclusion is strengthened by reference to the internal
documents of the parties. Part of the rationale for the merger is to achieve
a betterbalanced firm in the various international markets. 13 If there were
only a single international market there would be no such concern. Of note
as well is that the parties have different brands locally to those in the
offshore markets. This again is inconsistent with the notion of a single
international market.
34. In light of the above we find that the relevant geographic market for both
product markets can be divided into an international market, which itself is
composed of a series of offshore national markets, but this is not an issue
that we have to decide, and a domestic national market. For competition
purposes we will only consider the effect of competition in the domestic
national geographic market.
Market shares and concentration
35. The 2004 market shares of the 4 deciduous fruit canners in South Africa
are:
LFI 53%
RFF 20%
Ashton Canning 15%
Del Monte 12%
TOTAL 100%
36. In the fruit puree market the 2004 market shares are:
LFI 25%
Ashton Canning 25%
RFF 10%
Del Monte 25%
Boland Pulp 10%
record.
12 Record page 584.
13 See page 179 of the record.
8
Breede Valley Fruit Processors 5%
TOTAL 100%
37. The merged entity’s market share post the transaction would be 68% in
the national canned deciduous fruit market and 50% in the national fruit
puree market.
38. In terms of the oft quoted United States merger guidelines, utilising the
Herfindahl Hirschman Index (HHI), a post merger market of 1800 points is
considered highly concentrated, but that on its own is not decisive if the
change in concentration between the pre and post merger market figures
is less than 50 points; where in such a market the change is more than
100 points, it is presumed the merger will create or enhance market
power. 14 The HHI in the deciduous fruit canning market is set out in the
table below:
Canner % Market share HHI Premerger HHI postmerger
LFI 53 2809
4624
Ashton Canning 15 225
RFF 20 400 400
Del Monte 12 144 144
Total 100 3578 5168
39. In the fruit puree market the HHI is set out below:
14 See US Dept of Justice and Federal Trade Commission Horizontal Merger Guidelines as revised in
1997, paragraph 1.5.1
9
Canner % Market share HHI Premerger HHI postmerger
LFI 25 625
2500
Ashton Canning 25 625
RFF 10 100 100
Del Monte 25 625 625
Boland Pulp 10 100 100
Breede Valley 5 25 25
Total 100 2100 3350
40. As set out above it is clear that in the deciduous fruit and puree markets a
relatively concentrated market premerger with respective HHI’s of 3578
and 2100, became even more concentrated post merger with HHI’s of
5168 and 3350 respectively. The change in concentration was 1590 in the
deciduous fruit market and 1250 in the puree market.
41. Applying the US standard to the calculations in this merger, both the
canned deciduous fruit products and puree markets are, post merger,
highly concentrated and are likely to create or enhance market power.
42. Of course just as is in the US merger guidelines our approach to merger
evaluation is not to rest having found significant concentration, as there
may be other factors that prevail in a market that may detract from the
prima facie conclusion of the creation or enhancement of market power
that the concentration calculations suggest.
43. The factors we have regard to are enumerated in section 12(A)(2) of the
Act, which we consider below insofar as they are pertinent to this
transaction. We also, since the subsection list of factors is non
exhaustive, consider the Commission’s arguments around capacity.
10
Effect on Competition
Introduction
44. As we have discussed in the previous section the primary point of
departure between the merging parties and the Commission is over the
extent of the geographical market with the former contending for an
international market, while the latter opts for a national market. On the
other issues they are ad idem and this is because of the Commission’s
surprising conclusion that notwithstanding the fact that it finds this to be a
national market where post merger there is “from a structural perspective”
a substantial lessening of competition, the merger does not give rise to
competition concerns. 15
45. The Commission derives its comfort from two factors that it identifies in the
relevant product markets. The first is that it considers that retail chains are
able to exercise countervailing power over the merging firms. Secondly,
the Commission considers the existence of surplus capacity at the plants
of the other nonmerging canning firms, as a factor that will allow them to
discipline any attempt at market power by the merged entity. Although
neither of these two aspects was foremost in the minds of the merging
parties in their filings, they have, in their heads of argument, endorsed this
aspect of the Commission’s analysis.
46. We will consider both these arguments further below. We will firstly
consider the canned deciduous fruit market and then the fruit puree
market.
Canned Deciduous fruit market
Actual and potential level of import competition and ease of entry into the
market including tariff and regulatory barriers
47. Both the merging parties and the Commission place reliance on the fact
that imports place a restraint on the pricing behaviour of the merging firms.
48. According to the parties, low regulatory barriers, presently 5%, exist and
48. According to the parties, low regulatory barriers, presently 5%, exist and
imports can and do effectively compete with locally produced products.
They also stress that imports are capable of escalating dramatically at any
15 See Commission’s Heads of Argument paragraph 15 read with 22.
11
point in time. 16 After discussions with large retailers such as Shoprite
Checkers and Pick ‘n Pay, the Commission confirmed that retailers are
able to and do import canned deciduous fruit products, particularly from
Greece and China. It accordingly found that imports place enough of a
constraint on the merging parties from behaving anticompetitively.
49. Given the strength of the rand, which has had such a debilitating effect on
local producers in the export market, one would expect import competition
to be at its most competitive level now. However the evidence of Mr
Martalas is that presently imports constitute approximately 1% of the local
market. 17 This is hardly indicative of an actual or potential level of import
threat. As stated earlier the OFT considers a figure of at least 15% of
sales as an indication that exports are a factor in a domestic market. 18
50. There are several reasons why our market may not be attractive to
importers. In the first place our market is considered small and notably
even a locally placed producer in the form of Del Monte did not choose to
sell domestically until approximately 3 years ago. 19 Secondly, given the
competitive advantage that local producers enjoy as a result of their
location, product quality and strong loyalty towards some of the brands,
importers would need to be substantially cheaper to be able to compete.
Finally, importers would either have to develop a brand or can for a local
existing brand (such as one of those owned by a local firm) or a house
brand for a chain. Again we have no evidence that this has happened with
any significance premerger and it seems even less likely post merger,
given the enhanced market position of the merged firm and its ability to
command more than one brand.
51. Import competition is therefore unlikely to constrain an exercise of market
51. Import competition is therefore unlikely to constrain an exercise of market
power by the merged firm. We do accept the correctness of the parties’
assertion that tariff barriers are low but this does not detract from our
conclusion.
52. Nor is entry from a local firm likely either. According to the Commission’s
enquiries there had been no new entrant into the market since Del Monte
in 2002. The parties have not led any evidence to suggest that local entry
16 See parties Heads of Argument, par 18.
17 See record page 20. Martalas does suggest that once firms have got a foot in the door their market share
can “ multiply very quickly in a short period of time.” Competitors mentioned seem to be Chinese firms
who can apples. Beyond this, Martalas was rather vague as to who and how effective they were, despite
saying they were monitoring them. He did acknowledge that it was easier for them to get into the food
service market than the retail market as the former was less brand sensitive.( record 1819)
18 See footnote 6 supra.
19 See transcript dated 14 September 2005, page 200.
12
into this product market is likely. Nor could they have done so credibly
given the gloomy outlook for the industry which they so frequently express
not only during the course of this hearing but also in their public
statements and through their industry associations. Faced with the market
power of key suppliers, and having to negotiate competitive arrangements
with farmers, the prospects for a new domestic entrant are daunting and
unattractive. It also seems that bar entry from a competitor willing to break
the existing business model of the present incumbents, the new entrant
would have to ensure that the better part of its revenues came from
exports,; this given present exchange rates, coupled with the difficulty of
getting access to offshore markets is unlikely to herald entry.
53. Effective timely entry is therefore in our view unlikely given this scenario.
History of collusion
54. Our present canning industry was born out of the era of price and
marketing controls institutionalised by the control boards. Up until 1996
the price at which canning companies procured their fruit from farmers
was set by the Deciduous Fruit Control Board, as was the price at which
they exported their canned product. 20 In 1997 the control board was
abolished in terms of government policy to introduce market forces into
the agricultural industry. 21 Whether this was achieved in this industry is
open to some doubt. The Deciduous Fruit Control Board was replaced by
a voluntary association, the Deciduous Fruit Canners Association of South
Africa, which appears to replicate in function much of what the old Board
did.22
55. The associations do not fix prices. Instead what has replaced the Board’s
role in fixing the price of fruit at which farmers sell to the canners is a price
formula which each of the merging parties has for its respective suppliers.
formula which each of the merging parties has for its respective suppliers.
Briefly, the canning firm agrees to pay farmers according to a pre
determined formula based on the average selling price that the canner
relies on its sales in the export markets. Although Ashton and Tiger do not
apply the same formula, indeed Ashton’s seems more favourable to the
20 See record page 495.
21 For more background information on the Coops, see The Competition Commission v Patensie Sitrus
Beherend Ltd, Tribunal Case No: 37/CR/Jun01 and South African Raisins (Pty) Ltd and Johannes Slabber
v SAD Holdings Ltd and SAD Vine Fruit (Pty) Ltd, Tribunal Case No: 04/IR/Oct1999
22 In recognition of the specific needs of the deciduous fruit canners in the Western Cape, the Deciduous
Fruit Canners Association of South Africa was established under the umbrella of the SA Fruit & Vegetable
Canners’ Association, its largest members being Ashton Canning, RFF Foods and Tiger Brands. See page
502 503 of the Record.
13
farmers, neither has changed its formula in the past ten years. 23 Nor
according to Martalas do many farmers go from one canner to another
looking for a better formula; it is he said quite a stable relationship. 24
56. According to Martalas in his oral testimony:
“Those formulas have been in place for the best part of the last ten
years and it precedes my time in the industry, but they were as a
result of going from a regulated environment where the price was
set to a free and open environment. That’s where the formulas
were in place.”
57. In other words the demise of the control board and price fixing has been
replaced by an unchanging practice of long duration, symptomatic not of
the presence of vigorous market forces, but of the lingering ghost of price
control.
58. According to the testimony of Mr Wehan Victor the chief executive of the
farmers association, the Canning Fruit Growers Association, farmers
bargain annually with their canning firm customers. Yet even this
appearance of the working of a competitive market is not what it seems on
closer scrutiny. Mr Victor admitted that pricing formulas had not changed
in 10 years.
59. Indeed the meetings between the farmer representatives and their
canning firm customer serve more of an auditing function to ensure that
the average price to which the formula is to be applied has been applied
fairly. Noteworthy is that these meetings are not between individual
farmers, but their industry representatives in respect of that canning firm.
Farmers would appear to be price takers in the sense of the formula that is
applied although the average purchase price to which the formula is
applied is a function of international market forces in an aggregate number
of offshore markets in which the particular firm is exposed.
60. Since the exposure of each of the merging firms to international markets
60. Since the exposure of each of the merging firms to international markets
varied, we heard evidence that despite the fact that the formula applied by
LFI was less beneficial to farmers, their farmers on occasion reaped a
better return than their Ashton counterparts because LFI had realised
better prices in the overseas markets. (This again is evidence of the lack
of a uniform international price and hence that markets for these products
are national in character.)
23 See evidence of Martalas to this effect Record page 26.
24 Record page 27.
14
61. We requested at the prehearing, and subsequently received copies of
recent minutes of the industry associations namely the South African Fruit
& Vegetable Canners’ Association (Pty) Ltd and the Deciduous Fruit
Canning Forum. Whilst a fuller consideration of the purport of these
documents is not warranted the following features emerge. Meetings
between the competing firms in the industry are frequent and issues of
strategic importance appear to be discussed openly on a collegial basis.
Most canners share their expectations of pricing, costs, views on
innovation etc.
62. Whilst this may be more characteristic of agricultural product markets the
extent to which information and strategy is discussed between competitors
is again hardly suggestive of a culture of vigorous competition between
firms. Rather it appears that the firms see themselves as a national
collective, competing with other international players and hence
dissipating their possible national rivalry in favour of a common flag
rallying effort against foreign firms. The only firm to buck this trend has
been Del Monte a less than enthusiastic member of the club although it
appears despite this to recognise the self–interest in industry wide
collaboration.25
63. We would hasten to add that this is not to suggest that the canning firms
are engaged in activity that is unlawful. Rather what we seek to
emphasise is that the cultural milieu of the cooperative has not been
eliminated from the industry and it would be hard to regard this as a
market characterised by strong interfirm domestic rivalry. The merger
between the two strongest firms in the market can only serve to cement
this culture further. The concern expressed in the board minutes of Ashton
this culture further. The concern expressed in the board minutes of Ashton
that they would hope that if the merger were not to go ahead that this
might lead to an end to ‘friendly competition’ between the firms,
epitomises this culture.
Dynamic characteristics of the market including growth innovation and
product differentiation
64. This factor is primarily intended to account for the fact that in innovation
markets a first mover may have very high market shares but that should
not be suggestive of the fact that they will always be able to dominate the
market once new entrants have caught up. This does not apply here as
the goods are mainly commodities and there is little indication that Tiger
25 See transcript of 14 September on page 208 and record page 734 which refer to Del Monte’s attendance
at meetings of the Deciduous Canning Fruit Forum.
15
and Ashton’s positions of strength in the market are the result of new
innovations. On the contrary a reading of the minutes of the trade
associations shows how slowly our industry has been to innovate; the long
discussions over new packaging being one example of an idea discussed
by all competitors and yet to happen. Tiger as the leading firm has not
invested much and seems to have delayed any investment decision until
the industry had further consolidated. Thus the lack of competition in the
local market has seen a stultified market not characterised by either
innovation or, apart from Ashton recently, much investment.
Countervailing power
65. According to the evidence of Martalas most of the domestic sales of
canned deciduous fruit are to the retail market where most of the sales are
through the large retail chains, Pick n Pay, Shoprite Checkers, Spar or
through the wholesalers such as Metro and Massmart. A smaller portion
goes to what is termed the food service market, where the customers are
firms who are largely commercial caterers, such as Caterplus.
66. Insofar as the Commission spoke to retail customers they were of the view
that if the merged firm raised its prices they would be able to find other
substitutes including imports. On the face of it these are comforting
reassurances and seem to have given solace to the Commission. Yet on
closer inspection these are rote answers to a series of rote questions,
without any further evidence, analysis or even anecdote as to why this
countervailing power, if it exists, should be exercised. We have in previous
decisions expressed our reservations about considering retail chains,
despite their enormous bargaining power over suppliers of fast moving
consumer goods, as a font of countervailing power keeping down prices
consumer goods, as a font of countervailing power keeping down prices
for consumers. Rather, what the chains do is to extract discounts and
rebates from suppliers based on the list prices, so that the chains can be
competitive relative to one another in respect of a product; but as
suppressors of absolute prices i.e. keeping the list price down, there is
scant evidence that this is so nor any reason why they would want to do
so, since the retailer is not the ultimate consumer, but an intermediary
between the consumer and producer; it wants to retain its overall
competitiveness relative to its rivals over a product, but has less interest in
ensuring that the product, just one of the many it sells, is low in absolute
terms.
67. Nor has the Commission taken into account the fact that the merged firm
will be managed and twothirds owned by one of the largest suppliers of
16
food to the retail chains. 26 This firm is thus no skinny weakling falling over
in the face of the playground bully in the form of the retail chains; rather
the chains have an interest in stocking the portfolio of goods that Tiger
Brands can put on its shelves. This is not mere speculation. An
inadvertent addition to the record seems to confirm that this is so. The
Commission sought from Massmart answers to their various questions as
they had other retailers and received the customary answer, in
Massmart’scase in particularly terse and noncommittal language.
Interestingly however in sending its answer to the Commission, Massmart,
presumably unintentionally, enclosed some internal correspondence that
explains why the Commission received the type of answer that it did. The
author, one Graham Rebello who is the Forum Chairman of Massmart
Foods & Liquor Forum warns in an email written to subordinates who
would be supplying the answer to the Commission that: 27
68. “The detail required is not as thorough/complex as the Oil questionnaire
previously, but given that Tiger Foods Brands is involved in this merger,
Massmart is not going to respond in such a way that it may jeopardize this
merger being approved.”
69. This email suggests that not only are claims of the vast powers of the
chains to be treated with some caution when the supplier is a firm as
powerful as Tiger Brands or one of its subsidiaries, but also the reliability
of the views of chains who have no interest in prejudicing their commercial
relationships with powerful suppliers over the price of a can of peaches.
70. Countervailing power is therefore unlikely to curtail the merged firm’s
market power.
Failing firm
71. The merging parties at the outset of our hearing confirmed that they did
not intend to rely on the failing firm defence. 28 Despite this, a good deal
not intend to rely on the failing firm defence. 28 Despite this, a good deal
of evidence on this point has been led by the parties although in relation to
the public interest as the merging parties contend that absent the merger
there would be potentially disastrous effects on employment and the
economy of the Ashton region where the merging firms and the farmers
who supply them carry on their business. Although the parties seek not to
rely on the failing firm defence we will still consider this aspect.
26 Tiger states in its 2004 annual report that it gained the top three places, and five out of the top ten, in the
food sector, in a recent comprehensive survey of brands in South Africa conducted by Markinor. In
addition Koo emerged as the fifth most admired South African brand overall. See record page 209.
27 See page 476 of the record.
28 See record pages 56.
17
72. The merging parties tell us:
73. Ashton Canning’s bankers have expressed concern over the sustainability
of the business going forward. Ashton Canning forecasts a bank overdraft
of approximately R50 million for 2004/2005 and it anticipates that it will
require additional borrowings of approximately R100 million. It furthermore
projects a substantial net deficit in respect of the 2005 and 2006 financial
periods and should the current economic climate prevail, i.e. the strong
rand against the US dollar, Ashton Canning anticipates that its bankers’
reluctance to extend its overdraft facilities may well necessitate the
closure of Ashton Canning. This is particularly so when regard is had to
the fact that a rand/dollar exchange rate of approximately R7.25 to R7.50
is required in order for it to achieve breakeven point. With the current
exchange rate ranging from approximately R6.30 – R6.50 to the US dollar,
it will, accordingly, continue to record losses in respect of its business
operations.
74. Tiger too claims its LFI business is troubled. Since 2003 LFI has
experienced a loss varying from R4 million to R60 million with interest and
will, based on projections into the future, continue to do so until 2009.
Should these projections be correct, LFI, we heard, would be compelled to
rationalise its business to its most profitable core, which means it might
ultimately have to close its business. 29
75. Yet despite the gloomy picture painted by the figures it does not appear
that either firm is about to exit the market. We say this for two reasons.
The initiative for the merger came not from Ashton, supposedly the firm
most likely to fail, but from Tiger. It appears from the minutes that Ashton
was not the only firm with whom it considered a merger all were given
consideration and Ashton was considered the most suitable. Once Tiger
consideration and Ashton was considered the most suitable. Once Tiger
had made the approach Ashton whilst an interested party bargained hard.
It first resisted an outright sale of the business and demanded a share in
the equity of the merged firm, which eventually after much horsetrading
was settled on the present formula. Significantly the Ashton⅔ ⅓
minutes show that it considered as an alternative making an offer to buy
Langeberg from Tiger. This is hardly the behaviour of the management of
a firm about to fail.
76. This is perhaps the reason the parties wisely chose not to rely on the
failing firm defence. It must be pointed out that both in our law and in the
comparative case law, a failing firm defence will only succeed if the
29 See Tiger’s Heads par 5.2 on page 7.
18
merging parties can demonstrate that there is another buyer whose
acquisition of the target would raise less competition concerns. 30 In this
case as Tiger was the initial suitor we have no evidence that Ashton either
had sought a less objectionable buyer or indeed wanted to sell.
77. It should be noted that Ashton cried wolf previously in December 2003 in a
document it disseminated to interested parties at the time saying that the
“company was in real danger of being forced to close its factory unless
there is a steep weakening of the rand to the level of R7,50 to R8,00 to
the USD. The domestic market for canned fruit is very small and it is thus
not a viable alternative to switch sales to it .”31
78. Moreover Tiger, in its 2004 strategic plan, says:
“Notwithstanding the strengthening of the Rand over the past few
years, it can be argued that the long term agricultural economics
favour South Africa, especially against its EU and USA
competitors. Given the inherent quality advantage, South Africa
could become a world leader in canned peaches ”.32
79. The failing firm consideration is therefore not a factor we can consider in
relation to the competition assessment although we refer to this evidence
again when we deal with the public interest.
Removal of an effective competitor
80. At present Ashton is Tiger’s most effective competitor in the canned
deciduous fruit products market. Although we have two other firms in the
market neither is as effective a competitor as Ashton. Rhodes is the
smallest of the four incumbents, has the least spare capacity, and judging
from the evidence of Mr Bruce Henderson from Rhodes, does not see this
market as a priority as the firm is also involved in.markets for other canned
products; baked beans, vegetables and jams. LFI, in its Board briefing
paper, also mentions as one of the reasons why it had targeted Ashton
paper, also mentions as one of the reasons why it had targeted Ashton
Canning that Rhodes is “ an unwilling and in any event smaller player ”.33
81. Del Monte, whilst owned by an international food firm, is the least likely of
the four to remain committed to this market. Mr Henderson of Rhodes, in
30 See Iscor Ltd and Saldanha Steel (Pty) Ltd, Tribunal Case No: 67/LM/Dec01 .
31 See page 127 of the record.
32 See page 176 of record.
33 See record on page 179. [Confidentiality was claimed with regard to this document]
19
his oral testimony, described it as being “inconsistent”. 34 This is a
reference to the fact that the firm has experienced shareholder problems
in the past and has only recently started trading in the domestic market,
having previously been an exporter. Martalas also had a negative view of
Del Monte when explaining why it was not considered a favorable merger
partner:
“ I think there have been a lot of complications there in that group
over the last few years. The company effectively went into
liquidation in Europe and. It was bought out by another arm of Del
Monte in the United States. So there has been a lot in internal
turmoil.” 35
82. Del Monte has, it must be acknowledged, grabbed a good share of the
market in that short time although this appears to be because it saw the
domestic market as a balance to its offshore exposure since the rand
strengthened and has it seems been an aggressive price competitor, the
maverick of the local market. When questioned by the Commission it did
not appear that Del Monte had designs on much further expansion in the
local market. 36
83. Ashton by contrast is a stable and effective competitor despite its present
problems. In the first place it has been in the industry for many years; it
was established in 1949, and through all the cycles such an industry is
want to experience, has proved to be a resilient competitor. As a family
owned business it is run by a stable and experienced management, has
access to customers in a good spread of offshore markets, indeed it is the
envy of Tiger in this respect, and a good reputation for product quality. It
boasts an established client base in Europe, Far East, Middle East,
Canada and other markets. It states that some of its relationships with its
customers have lasted for more than 30 years. 37 In addition, the firm has a
customers have lasted for more than 30 years. 37 In addition, the firm has a
strong network of suppliers and has recently invested in new plant in its
cannery. According to Mr Du Plessis it has the best plant in the country at
the moment with a production capacity of approximately 44 600 tons of
fruit.38
84. The merger will lead to the number of competitors being reduced from four
to three, and whilst that in itself is a matter of concern, the greater problem
is that it sees the removal of LFI’s most effective competitor.
34 See transcript dated 13 September 2005, page 185.
35 See transcript dated 13 September 2005, page 14.
36 See transcript of 14 September 2005 on page 217.
37 See Ashton memorandum on the effect of strong Rand, record page 126.
38 See Competition Commission’s recommendation on page 8.
20
Capacity
85. The Commission places great reliance on the fact that if the merging
parties were to raise prices there is sufficient capacity in the rest of the
‘industry’ (for which read only Del Monte which has available access
capacity of 15000t during the peak season since RFF has none available)
for those firms to increase production and discipline the price rises
instituted by the merger firm. This theory was based on a questionnaire to
all the incumbents during the course of the Commission’s investigation. 39
But this theory has two flaws.
86. Firstly, capacity in this type of industry is dependant on both the size of
crops on the one hand and the ability to sell product in offshore markets.
The question was not whether there is spare capacity in general but
whether there is spare capacity during the relevant period namely the
canning season which lasts for six months of the year. During the season
it appears that there was little spare capacity and although after the
season there is, that is of no significance because it cannot be used to
increase production in this market where products perish if not canned
immediately. Secondly, and more significantly, neither of the two firms to
whom this theory must apply, if it is to have any credence, showed any
enthusiasm about expanding production for sale to the domestic market.
Indeed, as Mr Hofmeyer from Del Monte so frankly acknowledged, they
were worried that this would destabilize pricing in the domestic market. 40
87. It is highly unlikely that either of these two firms will aggressively pursue
expansion in the local market given that their strategy is to view the local
market as a currency balancing opportunity not a growth one.
Unilateral effects
88. Although Mr Martalas in his oral testimony testified that the rationale for
88. Although Mr Martalas in his oral testimony testified that the rationale for
the merger was all about efficiencies and denied that there was any
benefit to come from decreased competition, this is not borne out by the
internal documentation of the two merged firms. 41
89. In its strategic plan for its canned fruit business dated May 2004 Tiger
states:
39 See page 424 and 451 of record.
40 See transcript of 14 September 2005, page 218.
41 See Martalas on this point record page 17.
21
“It is a prerequisite that the cost basis is competitive against both
local and international competitors. This implies an everincreasing
focus on costs. The core of the business is commoditised and the
quality advantage of SA product only receives a 5 – 10%
premium…..The industry can rationalise further to a maximum of
three or even two manufacturers. Due to higher throughputs,
decreased competition and lower management costs the
profitability of the industry will thus increase.” (Our emphasis)
90. The same issue is also foreshadowed in the documents presented to the
Ashton board where among the reasons given in favour of the merger are
those that are described as “ Elimination of competition in export markets”
According to this document this would amount to R2R3 per carton. 42
91. In his oral evidence Martalas explained what this meant. Apparently both
Ashton and Tiger compete head on in international markets:
“Over the years obviously we find there is a bit of a price margin
erosion by just competing head on and we believe that by having a
single sales force and the different focus on markets, we should be
able to get better realisation.
92. And later on:
Ja, we also playing… its not just the customers are playing us. I
think we are kind of playing one another two, you know, kind of
desperation of getting the deal you’ll drop another 50c here and the
other person drops another 25c.” 43
93. Du Plessis in his oral evidence says much the same thing when he alludes
to the fact that the firms have some common customers who have “got
very sharp and they play us off against one another” 44 Du Plessis says
that this would contribute between 25 and 33% of the benefits. 45
94. Du Plessis testified that South African firms sell to international markets
from December to February each year. At that point in time only southern
hemisphere countries are producing and selling.
hemisphere countries are producing and selling.
42 See record page 182 , Ashton document headed Discussion points: Cooperation with Tiger Brands,
item A.1.
43 Record page 24.
44 Record page 82.
45 Record page 68
22
95. This then explains why the merged firms would be concerned about
eliminating competition between two of the South African producers
despite the fact that South Africa as a whole is only accountable for about
8% of sales internationally. Competition between Southern hemisphere
firms is fiercer because demand is seasonal although pricing is still a
function of overall nonseasonal supply and demand that includes
northern hemisphere producers. But what remains clear and the parties
are perfectly frank about this, is that part of the motivation for the merger
is to eliminate competition between themselves in relation to their
international customers and they believe this could gain them an amount
of [ ] 46 in savings if this price competition is eliminated. 47
96. The parties’ documents are silent on any strategies in the South African
market, which is not surprising given the low percentage of their business
it represents. Nevertheless if the parties are able post merger on their own
version to eliminate some price competition in international markets and
so sustain higher prices albeit of a low order, it is highly probable that they
could implement a larger price increase in the domestic market where they
will account for 68% of the market post merger.
97. It needs to be stressed that the firms do not compete in the local market in
the same way. Tiger primarily sells its own branded product KOO, while
Ashton owns no brands of its own, but cans for others; either third parties,
such as M&L or for supermarkets’ own labels or exclusive labels. For
instance the Ashton cans for a brand known as Farmgirl, which is a label
produced exclusively for Pick n Pay.
98. It is not clear how the merged firm intends dealing with brands other than
KOO in the domestic market going forward. They have indicated that they
KOO in the domestic market going forward. They have indicated that they
would be happy to continue canning for third parties post merger although
there is nothing in the documentary record to indicate that this has been
given any consideration. The fact is that even though Ashton does not
own its own brands this does not appear to be a factor that would
constrain a price rise by the merged firm to the third parties, who provided
they benefit from it would have no interest in constraining it. With the
merged firm’s percentage of the local market it is likely that it could
implement a small, but significant price increase without fearing that this
would be unprofitable. The two other firms in the market might well have
an interest in following them up. Certainly the evidence of Mr Hofmeyer
from Del Monte suggests that price stability in the local market is a matter
of concern to them. Rhodes seems to lack the inclination to expand much
46 Confidential information
47 See page 179 of the record. Document claimed confidential .
23
further in the local market given that canning is not its core business.
99. Furthermore from the pricing data we have received, Tiger and Ashton
already price at a premium to the other two firms, and certainly Tiger
prices at a substantial premium to Ashton which indicates that even pre
merger the firms’ are price makers. 48
100.Disturbingly in the shareholders agreement, which we only received after
the hearing, special provision is made in the list of prescribed matters that
both sets of shareholders have to agree to, for any decision that relates to
a decrease in volume of the fruit acquired by the company. Whilst one
does not want to read too much into such a provision, it does indicate that
the parties seriously contemplated that possibility, for it to find its way in as
a special protection.
101.We conclude that post merger the merged firm will unilaterally be able to
increase prices to consumers in the domestic market without such an
increase being unprofitable for.
Puree market
102.Thus far we have not looked at the other market implicated by the
merger, the food puree market. Although many of the same issues apply
to this market, as to the canned deciduous fruit products, there are some
differences, which lead us to the conclusion that in this market the merger
will not lead to a substantial lessening of competition.
103.The food puree market is a derivative market of the deciduous fruit
market. Fruit not suitable for canning is utilised in the lower value puree
market. In this market the merged firm will still have a very high share of a
highly concentrated market as our earlier HHI figures show. However the
concentration is less accentuated than in the canned deciduous fruit
products market. Furthermore there are more players in this market and
entry seems to be easier. According to the merging parties the capital
entry seems to be easier. According to the merging parties the capital
requirements to establish a fruit puree plant are approximately R20 million.
This is considerably lower than the cost of a plant to enter the deciduous
48 The comparisons between Tiger and Ashton come from material furnished by the merging parties
themselves as constituting list prices. The Commission obtained the prices from Rhodes and Del Monte
although in argument the merging parties suggest that the prices may not be comparable as they may not
include transport costs or provide for discounts. This may be so, but despite this, the price differentials
between Tiger and Del Monte are vast.
24
fruit market. 49 The time taken to enter this market is also less according to
the merging parties, twelve months as opposed to eighteen to thirty
months. 50
104.This submission on the ease of entry has been confirmed by the recent
and successful entry of Breede Valley, who in a short space of time had
acquired 5% of the market. Interestingly Breede did not consider that it
could attain the same success entering the canned fruit market. 51 Entry is
easier in this market because access to quality fruit is not an important
factor in competitive success as in the canned deciduous fruit products
market. Note that plant that is used in this market is not capable of being
used in the canned deciduous fruit products market. Thus competition in
this market is unlikely to be impaired post merger, notwithstanding the
high levels of post merger concentration, due to low barriers to entry
confirmed by the recent and successful entry of Breede Valley.
Will the merger lead to the creation of buyer power?
105.A number of witnesses expressed the concern that the merger would lead
to the merged firm having buyer power in respect of the farmers who
supply fruit to the industry. Most of the concern emanated from
competitors understandably worried that the merged entity could extract
better terms than they could. Farmers themselves were not consulted by
the Commission and for this reason we approached Mr Victor who as chief
executive for the farmers association is someone familiar with farmers
interests. He did not express great concern in this respect. 52
106.What is most probable post merger is that farmers producing for Ashton
will end up either supplying fruit on the more exacting formula that Tiger
gives to its suppliers, or a completely new arrangement will be negotiated
with all of them. The latter is the view expressed by Martalas, who, whilst
with all of them. The latter is the view expressed by Martalas, who, whilst
not giving any detail as to what this may be, expressed the view that
inefficiencies would need to be removed from the system. 53 It is probable
that the merged firm will be able to squeeze farmers further and that this
may see the exit of some producers. 54
107.However there is insufficient evidence to suggest that this will lead to
49 See parties merger fillings record pages 91 –92.
50 Record page 92.
51 See testimony of Mr Oelof Janse van Rensburg, transcript dated 13 September, page 105 and 112
52 See page 141 of the transcript dated 13 September 2005.
53 See transcript of 13 September 2005, page 27.
54 Although Martalas says that he intends to revisit the issue of farmer compensation he denies that they
will be squeezed further. Record page 28
25
buyer power over farmers. We do know from the evidence that farmers
can switch production out of canned deciduous fruit products to other fruit
or even other products. The opportunity cost of such switching has not
been modeled so we cannot meaningfully assess this. What we can
surmise is that if the merging parties squeeze the producers too hard, they
will exit, and the merged firm will lose its source of supply. Given that the
quality of our fruit is the greatest competitive advantage that our producers
have, this would be an ill advised strategy so we do not consider that an
anticompetitive squeeze is all that likely. 55 Accordingly although we find
that the merger will lead to the merged firm enhancing its bargaining
power with farmers we find that will fall short of a form of buyer power that
is anticompetitive.
Conclusion
108.The merger will lead to a highly concentrated market in the deciduous
canned fruit product market where market power may be created or
enhanced. Whilst high concentration levels are not dispositive of whether
a merger will lead to a substantial lessening of competition, the other
factors that we have considered seem to confirm, rather than detract from
that conclusion; new entry into the market is unlikely in the foreseeable
future, the market itself has not been characterised by great rivalry and is
likely post merger to see a continuation and solidifying of the collusive
tendencies already manifest; even absent collusion the merged firm will be
able to wield market power unilaterally; the merger will also see the
removal of an effective competitor probably the firm best placed to
challenge Tiger in the local market; countervailing power and excess
capacity are not credible threats to the merged firm’ s market power.
capacity are not credible threats to the merged firm’ s market power.
109.Accordingly we conclude that the merger will lead to a substantial
lessening or prevention of competition in the national domestic market.
Efficiency defence
110.In terms of section 12A(1)(a)(i):
1) Whenever required to consider a merger, the Competition Commission
55 See copy of an extract from the article by David Kaplan and Raphael Kaplinsky: Trade and Industrial
Policy on an Uneven Playing Field: The Case of the Deciduous Fruit Canning Industry in South Africa, on
page 590 of the record.
26
or Competition Tribunal must initially determine whether or not the
merger is likely to substantially prevent or lessen competition, by
assessing the factors set out in subsection (2), and
(a) If it appears that the merger is likely to substantially prevent
or lessen competition, then determine
(i.)Whether or not the merger is likely to result in any
technological, efficiency or other procompetitive gain
which will be greater than, and offset, the effects of any
prevention or lessening of competition, that may result or
is likely to result from the merger, and would not likely be
obtained if the merger is prevented.
111.Whilst not specifically invoking an efficiency defence, because their case
has been hinged on a finding that the market is international, the merging
parties have nevertheless alleged that the merger will lead to certain
efficiencies, and we must consider whether they can resurrect the merger
given our conclusion that the merger will lead to a substantial lessening of
competition in the canned deciduous fruit products market.
112.As the subsection from the Act we have quoted indicates, it is not
enough for the merging parties to allege that there will be efficiencies, post
merger. The efficiencies to trump the anticompetitive effect must –
i.)outweigh the anticompetitive effects of the merger; and
ii.)be obtainable as a result of the merger.
113.We have previously held that merging parties bear the onus of proving the
efficiency defence. 56
114.What the merging parties have argued is that the merger will lead firstly
to certain production efficiencies as a result of the firms being able to
integrate their production lines and to shorten the production period.
Secondly, they allege that they will be able to increase investment.
115.The most detailed documentation for the efficiency claims comes from a
115.The most detailed documentation for the efficiency claims comes from a
document prepared by the Tiger board for its deliberations on whether to
conclude the merger. In this document efficiencies in the amount of [ ] 57
are alleged and should it be achieved, would result in a reduction in their
56 See Tribunal Case No: 89/LM/Oct00, Trident Steel (Pty) Ltd and Dorbyl Ltd for the acquisition of three
operations of Baldwins Steel, a division of Dorbyl Ltd.
57 Confidential information .
27
breakeven point from [ ] to below [ ], i.e. it would remain viable if there
was a 12% decline in the relative exchange ratio. 58
116.We went through these items with Martalas in his oral testimony and what
emerged was that a good part of the efficiencies claimed are as a result of
the shorter production period. 59
117.Some understanding of the production constraints of a canning firm is
needed to appreciate this. In South Africa the canning season lasts from
November to May the following year. During this season canning firms can
four types of fruit. They commence with apricots in November and then in
consecutive order, to peaches in January, pears in March and apples in
May. To reduce its costs the factory wants to do two things. Firstly it wants
to reduce production time. Production is labour intensive and involves the
mass employment of low paid seasonal workers who are employed for as
long as the season lasts, thus, the shorter the production run, the less the
expense. Secondly, because fruit degrades so quickly it is important to be
able to can it as soon as it has been harvested. Since canning requires
the highest grade of fruit and there are higher margins earned on canned
fruit, canners are anxious to optimise their purchase before it degrades
and can then only be used in puree manufacture, a lower value added
product.
118.Whilst the deciduous fruits that are canned in the region largely follow
seasons that are consecutive, there are some overlaps when the cannery
will be completing one fruit and receiving the next. Because of the
vicissitudes of nature, the yield of fruit and the factory’s production needs
are often out of alignment. One crop may be large and late, while its
successor is large and early. A factory with only one production line will
end up wasting fruit from both crops. The merger will thus assist both firms
end up wasting fruit from both crops. The merger will thus assist both firms
to use their production lines to specialise in a particular fruit until the run is
ended thus reducing the time to process the fruit and to reduce the
degrading factor.
119.The other advantage of having two production plants is specialisation
within a fruit category. Some markets such as Japan require products of
high grade and pay the premium accordingly, whilst others are happy with
a lower priced and lower quality product. Here again having two facilities
allows the merged firm to run the different quality fruits on different lines,
which again leads to greater efficiency.
58 See confidential document on page 179. Figures in brackets claimed confidential.
59 According to the parties the continuous processing of fruit type would enable the high season period to
be shortened by approximately 3 – 4 weeks.
28
120.What makes this type of efficiency possible is the physical location of the
plants. Ashton Canning and LFI are across the river from one another.
This also explains why investment in further plant enhancing facilities such
as a labeling plant makes sense. At the moment, in the Tiger documents,
this manufacturing efficiency will yield an amount of [ ] 60 excluding the
saving on reduced labour, an amount of [ ] 61. It is the former saving
that we are interested in.
121.The other cost reductions that go into the [ ] 62 that Tiger claims the
merger will realise are less compelling on further inspection.
122.The first amount claimed, [ ] 63, represents the increase in margin that
the merged firms would earn if there was a reduction in competition in the
export sector between themselves. This represents the quantification of
the anticompetitive gain in the export market and is certainly not
cognizable as an efficiency gain. The next category come in the form of
cost reductions from suppliers, growers and can manufacturers, as the
merged firm will be able to benefit from the greater discounts which Tiger
was able to extract from suppliers because of its other food operations,
discounts that Ashton Canning was unable to command. Further there are
the reductions in labour costs, namely the cost of having less seasonal
workers in the aggregate workforce of the merged entity due to the
shortening of the production process.
123.Thus we are left with only one set of efficiencies that does not come at the
expense of overseas consumers, or another player i.e. worker or supplier,
what are termed pure pecuniary gains. This is the [ ] 64, which
represents a dynamic production gain, and whilst not vast, is at least
credible, conservative and ongoing.
credible, conservative and ongoing.
124.Whilst the language of the Act does not distinguish between types of
efficiencies we have, in Trident Steel , held that this is relevant to the
question of whether the Act requires there to be a showing that efficiency
gains or part thereof will be passed on to consumers in the form of lower
prices. We held in that case, whilst such a showing is not fatal to an
efficiency defence, we would consider efficiencies as a continuum so
where there is no pass through demonstrated we would be stricter about
60 Confidential information.
61 Confidential information.
62 Confidential information.
63 Confidential information.
64 Confidential information.
29
the types of efficiency we would recognise. 65 Thus, at the highest end of
the continuum were dynamic efficiencies such as innovation whilst at the
lowest are the pure pecuniary efficiencies.
125.In this merger the parties have made no showing that cost reductions will
be passed on to consumers either domestic or international. Rather it will
improve the margin of the merged firm to return it to levels enjoyed when
the rand was weaker. Absent this showing we will view the efficiencies
claimed more strictly and find that only the [ ] 66 bears scrutiny.
126.Does this amount outweigh the loss to competition that we have found
that the merger will bring? We answer this in the affirmative. We say this
because although we recognise the efficiency gain against the merged
entity’s entire production run, both domestic and international, we
measure the anticompetitive effect only in the South African market, as
our jurisdiction in terms of section 3 of the Act is confined to economic
activity having an “effect within the Republic” The bulk of the anti
competitive effect in this merger will be felt in export markets. We do not
have a figure quantifying the anticompetitive effect. It is, however, likely to
be confined to the domestic production represented by Ashton, and if we
are correct about the possibility of a coordinated effect, the production of
the two remaining firms, Rhodes and Del Monte. We say this as it highly
unlikely that the Koo brand can be priced any higher than it is presently.
Already this brand is priced at an 18% premium to that of Ashton. We
cannot be certain of where Ashton is priced relative to its other two rivals.
127.Although the Commission has obtained pricing information from them per
carton, the merging parties suggest that the information is unreliable, as
we do not know what other costs are excluded, such as transport;
we do not know what other costs are excluded, such as transport;
secondly, that in the retail market, lists prices mean very little as in reality
all pricing is influenced by either discounts off list prices or rebates. When
asked about pricing relative to one another during the hearing none of the
witnesses were that forthcoming, indeed some answers were
contradictory and thus we cannot get a reliable sense of what is
happening in the market. We can however at least conclude that Rhodes
and Del Monte price well below that of Koo, and that unless their list prices
are wholly unreliable, also price below Ashton, if only marginally. A post
merger price increase would therefore be felt across the sales of all three
firms, but is unlikely to be substantial.
128.Given that the efficiencies here have been pared down to the most
65 See Trident Steel case supra.
66 Confidential information.
30
compelling, and that those are taken over 100% of the production of the
merged firm, measured against the anticompetitive loss over 15% of the
production sold domestically, or more likely less post merger if market
power is exercised, we can assume that at least qualitatively, the
efficiencies are, going forward, likely to outweigh the anticompetitive
effect even on a coordinated effects outcome.
129.We find therefore that it is most likely that the procompetitive gains from
the merger would outweigh the potential loss to local consumers.
130.The production efficiency we have shown is merger specific and hence is
cognisable. It is therefore not necessary for us to consider the investment
claims that the parties make that they will enter into post merger although
we have doubts as to why this investment is merger specific and could not
be attained without the merger. 67
131.We find that the merger will lead to an efficiency gain, which will be
greater than and offset the loss to competition that comes about as a
result of the merger.
Public interest
132.The public interest proved to be the most contested terrain between the
merging parties and the Commission. The Commission has contended
that the merger will have an adverse effect on employment, in that on the
parties own version, the merger will lead to an employment loss of 45
permanent jobs and 1000 seasonal jobs. For this reason the Commission
has sought to impose a condition on the merger in respect of employment
loss, the gist of which is that the merging parties should set up a training
fund that would not only benefit retrenched workers but any other member
of the Ashton community. The Commission states that the merging parties
should settle an amount of R2 million on the fund and that they subject
themselves to a reporting regime to report on the implementation of the
themselves to a reporting regime to report on the implementation of the
fund.
67 The claim that the merger will lead to higher investment is problematic from another point of view.
Whilst not doubting the parties’ bona fides in this respect this claim not only lacked any detail no business
or investment plan was produced, but it also failed to cross the most important hurdle that this was merger
specific. Nothing stopped either firm investing in plant premerger and although pressed on this point
Martalas was not able to explain apart from a preference for not investing without a merger why this could
not take place without it. Doubtless Tiger Brands does not want to invest in plant without a greater share of
the market and certainly a more stable environment for which read less of a competitive threat before it
commits to investing further in the present unfavourable conditions, but that does not make an investment
decision merger specific. Convenience does not translate into making such efficiencies, whatever they may
be and the merging parties do not tell us, merger specific.
31
133.The merging parties invoke the public interest in three ways, the effect the
merger will have on the region, on employment and on SMME’s. 68 Recall
that they have said that without the merger the business of Ashton is likely
to fail while the business of LFI will have to be scaled back. This they say
will impact negatively on employment, because although the merger will
lead to an employment loss of 45 permanent jobs and 1000 seasonable
jobs, without the merger the employment loss will be much greater if
Ashton closes; approximately 87 of LFI’s and 186 of Ashton Canning’s
permanent employees will lose their jobs and, further, approximately 1000
of LFI’s and 3000 of Ashton Canning’s seasonal workers will lose the
possibility of employment. 69
134.They also state that if Ashton closes this will have an effect on SMME’s in
the area who supply it whilst LFI would scale back its purchases. The final
public interest ground that the parties rely on is that the merger will
“enhance the ability of national industries to compete in international
markets”. In view of the fact that we are only now concerned with the
question of whether the public interest does not justify the merger we need
not consider this latter argument any further.
135.The parties’ employment and SMME case is, as we have seen, largely
premised on the idea that Ashton will, sans merger, fail and that LFI will
drastically cut back. What this amounts to is arguing a failing firm defence
though the back door when it was unsustainable through the front door.
Just as the failing firm defence requires the merging firms to show that
there is no more preferable buyer for the merging firm, so under the public
interest they need to show that no one else would be willing to buy Ashton
if it failed. This they have not even begun to demonstrate.
if it failed. This they have not even begun to demonstrate.
136.Although the parties state in their heads of argument that there is
“compelling and uncontroverted evidence that Ashton Canning itself
anticipates such closure...” this is simply not correct. 70 Whilst that may
have been the oral testimony of Mr Du Plessis, his documents prepared
for the Board concerning the merger show no such dramatic concern.
Indeed there is no mention of the fact that if the merger does not proceed,
the alternative is closure. 71 While the minutes consider options open to
Ashton if the merger is not consummated there is no mention of this
dismal prospect.
68 See section 12A(3)(a),(b) and (c) of the Act.
69 See parties Heads of Argument on page 37.
70 See merging parties Heads of Argument paragraph 52.
71 See Board Briefing Paper, page 182 of the record. [ Document claimed confidential ]
32
137.We find Ashton’s oral testimony in this regard unreliable when measured
against the documentary record on which we place greater reliance. Nor is
this claim of imminent failure credible, given Ashton’ s robust approach to
the negotiations and its consideration of buying LFI as an alternative if it
could not get commercially what it wanted. In the same way we rejected
the failing firm possibility in considering the competition issues, so we
reject it here. That is not to say that if the merger is not proceeded with,
that there may not be employment consequences if the firm’s scale back
production, but there is no evidence on record that this has been
contemplated and that this would dwarf the employment consequences of
those employment losses that the merger will bring about.
138.We now consider whether the merger itself will lead to employment losses
that constitute a substantial public interest concern. The Commission
states that the merger will have a net negative effect on employment and
hence the condition it imposes. Note that FAWU also objects to the
merger and points out that the merger envisages a loss of 1015% of the
permanent employees and 1520% of the seasonal employees. 72
139.The numbers involved are common cause and come from the merging
parties. Unlike the employment effects of the firms if the merger was
prohibited the post merger employment losses have been documented by
the parties and quantified.
140.Out of a combined work force of 447, 45 permanent employees would be
affected. Out of a combined average annual employment of 14 439
seasonal workers at least 1000 less will be employed each year.
141.Seasonal workers are persons who annually are employed in the industry
for short periods of time during the high season. The high season lasts for
a period of six months, but not all seasonal workers are employed for the
a period of six months, but not all seasonal workers are employed for the
full season. Martalas testified that the more labour intensive crops come at
the beginning of the season and thereafter there is a gradual drop in the
employment of seasonal workers so that on average only 5060% of those
who commence the season are still employed at the end. 73
142.Although seasonal workers are only considered employees for the
duration of their seasonal contracts there is a practice that people
continue their employment with the cannery the following season.
Martalas admitted that some of the seasonal workers had served more
72 See page 391 of the record.
73 See transcript dated 14 September 2005, page 237.
33
than fifteen years with the company. It is not clear what seasonal workers
do when they are not employed during the season, we asked Martalas this
and he did not know. What is known is that seasonal workers are unskilled
workers and that if 1000 less are to be engaged by the merged firm post
merger, their prospects for other employment are limited. It is common
cause that Ashton is heavily dependant on the canning firms since it is an
economically troubled area that offers little hope for unskilled labour.
143.Given that this large number of workers will, post merger, have their
employment possibilities seriously reduced and that they are not likely to
gain employment elsewhere, we find that the merger will have a
substantially negative effect on employment and hence the public interest.
144.We now consider the debate between the parties and the Commission on
the condition that the Commission has proposed in its recommendation.
The debate is not that extensive. Whilst the merging parties would like the
merger cleared without any condition, they did tender a proposal that the
Commission has recrafted in albeit from the parties perspective in a
manner that they find too onerous, both financially and administratively.
145.The merging parties had offered an amount of R250 000 and consider
the amount of R2 million as excessive. They also have difficulty with the
Commission’s wish that the range of beneficiary who could benefit from
the fund be extended quite broadly to include “ the unemployed Ashton
community at large, including the seasonal workers of both merging firms ”.
146.We agree with the merging parties that to increase the range of potential
beneficiaries in the manner that the Commission proposes would be
tantamount to turning the merged firm into an enlarged welfare agency. It
is also in the context of a deprived community likely to be a vehicle for
is also in the context of a deprived community likely to be a vehicle for
conflict as well as rendering ineffective precisely what the Commission
seeks to achieve – to make individual contributions more meaningful.
147.We accept that the class of beneficiaries should be confined to those who
have experienced the employment effects most directly, retrenched
employees and seasonal workers who have worked in the past and who
have not had their contracts renewed by the company.
148.We find sympathy with the Commission’s criticism of the offer of R 250
000 as inadequate. Given that there are to be a potential class of at least
1000 people this would amount to each individual receiving an amount of
R 250. As a contribution to their retraining, the avowed purpose of the
scheme, it is as wanting in utility as it is in generosity. The parties plead
34
that as the businesses are presently both unprofitable and will not be in
the near future that this is what they can afford. The Commission correctly
points out that measured against the claimed total efficiency gain per
annum this once off payment is not excessive.
149.We agree. Having found that the merged firm’s efficiency claims are
credible we have little doubt that they will be in a position to make this
once off contribution in the amount of R 2 million, without serious prejudice
to their business, especially as we have found that shareholders will be
the beneficiaries of these gains, as we have found that they are not likely
to be passed on to consumers. Such a claim of impoverishment ill befits a
firm that has, post merger, pretensions to be a national champion, let
alone a firm that boasts one of the country’s major conglomerates as its
twothirds shareholder.
150.We find that the amount of R 2 million is reasonable to both the merged
firm and to the class of beneficiaries, indeed, if anything, it errs on the low,
not the high side. If the contribution was any less it would fail to be useful
as retraining opportunity for those affected.
151.We have therefore imposed a condition that builds on the foundations of
both the drafts of the merging parties and the Commission. We have
limited the class of persons to former employees. We have also limited
this further by excluding management employees who will be more able to
find reemployment. We have however extended the period of moratorium
on retrenchments and the ceiling on the reduction of seasonal employees
for a period of three years after this order. Note that the limitations that we
have placed are not ones that we have contrived – they are what the
parties state will be the likely employment effects of the merger. Nor does
parties state will be the likely employment effects of the merger. Nor does
this limitation affect nonmerger specific retrenchments or employment
reductions.
152.Having found that the merger will have and adverse effect on
employment we are satisfied that the condition that we have imposed will
in some measure serve to ameliorate them.
153.The merger is accordingly approved subject to the condition contained
below.
35
____________
N Manoim Date
Concurring: Y Carrim and M Holden
ANNEXURE A
Conditions
1. During the period the merged entity is subject to the following employment
limitations:
a. It may not retrench more than forty five (45) employees from the
aggregate number of employees employed by both firms
immediately prior to this order; and
b. Its average employment of seasonal workers during the period
should not be one thousand (1000) less than the average
aggregate number of seasonal workers employed by both the
merging firms, calculated over the three high seasons that
immediately preceded this order.
2. The permissible retrenchments in 1(a) or employment reductions in 1(b):
a. may take place at any time during the period; and
b. do not preclude the merged firm from undertaking further
retrenchments or employment reductions during the period
provided they are not merger specific.
3. Within thirty (30) business days of this order the merged firm must provide
36
the Commission with the figures for
a. the average aggregate number of seasonal workers employed by
both the merging firms calculated over the three high seasons that
immediately preceded this order; and
b. the aggregate number of employees employed by both firms
immediately prior to this order.
4. The merged entity must annually within thirty (30) business days of the
end of each high season, during the period, provide the Commission with
the details of any retrenchments or employment reductions undertaken
that year, regardless of whether they are merger specific and the reasons
therefore. For the purpose of this reporting obligation, the year will run
from 1st May the one year to 30 April in the following year.
5. The merging parties will make available an amount of two million rand (R2
000 000) (“the designated training amount”) for the purpose of training all
effected persons.
6. Any effected person wishing to avail himself or herself of any training shall
make application to the merged entity for the allocation to him/her from
the designated training amount of all or a portion for the fees payable by
him/her for the purposes thereof. Such application shall be made on a
fully motivated basis and shall include details (on accredited
documentation of the training facility in question) of the cost of the training
programme, the material covered in the course and the certification or
other accreditation conferred on successful participants upon completion
of the training programme.
37
7. The merged entity shall have the discretion to determine how much each
effected person shall be entitled to be allocated from the designated
training amount, but subject to the following principles:
a. Insofar as it is possible the designated training amount should be
allocated equitably among effected persons who apply;
b. The designated training amount must be fully disbursed to eligible
persons within thirty (30) business days of the end of the period;
and
c. No amount need be allocated to an effected person who does not
apply in the manner contemplated in clause 6.
8. The parties will on a quarterly basis, submit to the Commission a schedule
indicating those persons on whose behalf monies have been advanced
and the quantum thereof and those persons (if any) whose applications
have been refused. The reporting obligation contemplated shall
commence upon the expiry of the first quarter following this order and
shall endure until thirty (30) business days of the end of the period.
9. For the purpose of this order:
“effected person” means any person either retrenched in terms of 1(a) or
the subject of an employment reduction in terms of 1(b), but excludes
management level employees;
“high season” means the period from 15 November in one year to 30 April
in the following year;
“period” means the period from the date of this order until the 30 April
2008.
38
39