COMPETITION TRIBUNAL
REPUBLIC OF SOUTH AFRICA
Case Nos: 33/LM/May02 – 38/LM/May02
In the large mergers between:
Sandown Motors Holdings (Pty) Limited and McCarthy Limited
and
Sandown Motors Holdings (Pty) Limited and Barloworld Motor (Pty) Limited
and
Barloworld Motor (Pty) Limited and Durban South Motors (Pty) Ltd
and
Newco, (being a joint venture company between Barloworld Motor (Pty)
Limited and Durban South Motors (Pty) Ltd) and McCarthy Limited
and
Sandown Motors Holdings (Pty) Limited and Imperial Holdings (Pty)
Limited
and
Imperial Holdings (Pty) Limited and Sirius Motor Corporation (Pty) Limited
________________________________________________________________
Reasons for Tribunal’s Decision
________________________________________________________________
Introduction
1
1. We have been asked to approve six transactions 1 involving the sale of
firms who are authorised dealers for the products of Daimler Chrysler
South Africa (‘DCSA’).
2. Ordinarily, each of these mergers would be considered separately since
legally, at least, they constitute discrete transactions. However, with good
reason, the merging parties and the Competition Commission have
treated them as one for the purposes of their evaluation. We have been
commended to follow that approach which we have done.
3. The reason for this approach is that the six transactions all involve the
same four firms, variously cloaked in the garb of buyer and seller, and are
driven by the same rationale; the desire by DCSA to implement its
ambitious dealer network strategy (‘the DNS’).
4. In a nutshell, what is happening is that DCSA has reorganised its sales
network in certain urban areas, into five geographical territories, which it
refers to as “metro centres”. Only one firm will be awarded a franchise for
a particular metro centre. Each of the four firms of dealers will be allocated
at least one territory, whilst one firm will be located two. Since presently
some of the firms are located in more than one territory it was necessary
to engage in the several transactions before us to effect the re
organisation.
5. Post the series of mergers, the DCSA dealer network in the five metro
areas will in the words of the DNS be characterised by “bigger territories
with less owners” 2We have to decide whether the transactions in question
will substantially lessen or prevent intrabrand competition, and if they do,
then the question is whether there remains sufficient interbrand
competition to ameliorate any concern about the possible loss of intra
brand competition.
6. Although the mergers in question are horizontal in nature, they form part
6. Although the mergers in question are horizontal in nature, they form part
of the manufacturer’s strategy to realign its own distribution network this
means that the competition concerns, if any, should be examined from a
vertical rather than horizontal perspective.
History of the Matter
7. The mergers were notified to the Commission in May 2002. On the 31 July
2002 the Commission recommended in terms of section 14 A of the Act
1 Five horizontal mergers and a joint venture agreement.
2 Page 24 of the DNS document.
2
that they be approved without conditions. The Retail Motor Industry
Organisation (RMI) 3, an industry body that represents dealers, applied to
intervene in the proceedings. There was no objection from the merging
firms and we allowed them to do so on the basis of their interest in the
matter. The RMI argued that the merger should be approved, but subject
to conditions relating to the franchise agreement between DCSA and its
dealers.
8. On the 14 August 2002 we held a prehearing conference and requested
further documentation from the parties.
9. The matter was heard on the 11 September 2002. Apart from hearing
submissions from the parities, the Commission and the RMI, the parties
also led oral testimony from two employees of DCSA South Africa, Mr
Christoph Kopke, the Chief Executive Officer, Fritz van Olst, the Sales and
Marketing Director, as well as Mr. Phillip Michaux, the Managing Director
of Cargo Motors, who also serves as the chairman of the Mercedes Benz
Dealer Council 4.
10. On the 12 September 2002 we ordered that the merger be approved
without conditions.
Background to the DNS
11. Daimler Benz recently merged with the Chrysler Corporation in one of the
largest corporate mergers ever, to form Daimler Chrysler the parent of
DCSA. The merger greatly increased the portfolio of products to be
distributed by the merged firm, but also meant that the company was now
distributing very different product offerings. 5 This issue has vexed DCSA
considerably as it meant that in the same outlet a number of unconnected
or competing brands appeared under the same roof. In the minds of their
marketing people, this was detracting from the brand equity of their
premium marque, Mercedes Benz. Put in the words of their CEO one
would find a Colt bakkie next to a Mercedes SLK on the same showroom
would find a Colt bakkie next to a Mercedes SLK on the same showroom
floor. The proximity of the plebeian cart to the patrician coach would serve
to diminish the latter’s value in the eye of the beholder. Thus an important
component of the DNS is to end multibrand DCSA show rooms and
3The RMI comprises 6,500 members with branches in six cities. It represents 95% of all
franchised motor dealers within South Africa.
4 The Dealer Council is elected from the dealer network of 100 DCSA dealers and it meets with
DCSA management board on a quarterly basis.
5 Previously, as Mercedes Benz South Africa Limited, the firm had also been responsible for the
distribution of Honda products but claims that these products were complimentary to, and not
competitive with its other offerings. ( see Competitiveness report page 1)
3
replace them with those dedicated to their specific brands, namely
Mercedes Benz, Chrysler, Colt and Mitsubishi.
12. But the redesign in strategy that is sought in the DNS does not end there.
Another major aspect is to change the focus of its dealers. It is this aspect
of the strategy which is perhaps the most contentious and for this reason
we need to examine the status quo first before we deal with the new
proposals.
13. Various models for motor vehicle distribution exist and our market has
examples of all of them. In the first place we have dealerships that are
owned by the manufacturers. This apparently is the norm for the sale of
commercial vehicles and has been known, although is not the norm, in the
distribution of passenger vehicles. 6
14. This is because the distribution of passenger vehicles requires a
ubiquitous network, to make a manufacturer attractive to consumers, and
hence a high level of investment downstream, which is not something
manufacturers would readily assume.
15. This has led to dealerships being undertaken by firms owned
independently of the manufacturer who are either dedicated dealers of
that manufacturers’ products, or distribute the products of several
manufacturers, the so–called “multi– brand” franchises.
16. Nevertheless even multi–brand franchisees have not distributed more than
one manufacturer’s products in any one outlet. Thus, by way of example,
although the McCarthy Group holds franchises from most of the major
manufacturers, it has dedicated outlets for each manufacturer.
17. In the second hand market things are different and it is not
unusual to see the products of rival manufacturers side by side
on the same show room floor. Why has this distinction come
about? It is because one cannot become a retailer of new cars
about? It is because one cannot become a retailer of new cars
without a supply agreement with a manufacturer, which will
invariably, as one of its conditions of supply, require the dealer
to provide a dedicated show room for its products. 7
6 This was the approach formally adopted, although since abandoned, by BMW. It is also a
centrepiece of DCSA’s strategy through their acquisition of a 75% stake in Sandown Motors, as
appears more fully below.
7 In response to a question posed at the hearing by the panel with regard to whether
dealers could have a giant showroom from where they would market a number of competing
4
18. In the second hand market a dealer does not need any contract with a
supplier and hence the condition is not practically enforceable.
19. The final species of dealer is the exclusive dealer who retails only the
products of a specific manufacturer but is owned by a firm independent of
the manufacturer but as with the multibrand dealers, has a franchise
agreement with the manufacturer.
20. In 1999 DCSA, as part of the evolution of the dealer network strategy, took
a strategic decision to enter the dealer network directly and hence took up
a share in one of its exclusive dealers, by acquiring a 75 % stake in
Sandown Motors.
21. The merger was notified to us as a large merger and we approved it
without conditions. 8 Already at that stage there were murmurings of
unhappiness amongst the dealers and the RMI initially indicated that it
wanted to intervene in our proceedings to oppose the merger, but later did
not do so, formally withdrawing their objection to the merger at the
hearing, after a memorandum of understanding was reached in terms of
which DCSA undertook to maintain transparency and to consult with the
RMI with regard to all aspects of its new strategy in an appropriate forum.
The RMI were nevertheless invited to make submissions with regard to
various aspects of the transaction, as well as to the nature of the industry
in general.
22. DCSA as it was entitled to, has proceeded to implement the merger with
Sandown Motors. The present six transactions represent the next phase
of the implementation strategy.
The Transactions
23. Sandown Motors Holdings (Pty) Limited and McCarthy Limited
SMH will acquire from McCarthy its retail motor outlets in Randburg,
Milnerton, Claremont and Culemborg. McCarthy will acquire from SMH its
Milnerton, Claremont and Culemborg. McCarthy will acquire from SMH its
Pretoria outlets, being Ellenby Motors and its Mitsubishi dealership.
24. Sandown Motors Holdings (Pty) Limited and Barloworld Motor (Pty)
Limited SMH will acquire from Barloworld its passenger car and
brands, Mr Van Olst replied:“ The current franchise agreements, I think, from all the
manufacturers would prevent that. That is not, they can’t choose to do that .”
8 Our decision in this matter is reported as DaimlerChrysler SA (Pty) Ltd and Sandown Motor
Holdings (Pty) Ltd under 44/LM/Jul01.
5
commercial vehicle outlets in Roodepoort, being Garden City Roodepoort
PC and Garden City Roodepoort CV.
25. Barloworld Motor (Pty) Limited and Durban South Motors (Pty) Ltd (the
joint venture)
The parties are to enter into an agreement in terms of which they will set
up a joint venture with the view of combining their motor retail outlets for
DCSA products in the Durban area. Barloworld Motor (Pty) Ltd is to
contribute its dealerships in the area to the joint venture, whilst Durban
South Motors (Pty) Ltd will contribute its dealership in the area to the joint
venture.
26. Newco, (being a joint venture company between Barloworld Motor (Pty)
Limited and Durban South Motors (Pty) Ltd) and McCarthy Limited
All parties to this transaction are retailers of motor vehicles. The parties
are to enter into an agreement in terms of which the joint venture between
Durban South Motors and Barloworld Ltd is to acquire from McCarthy its
passenger cars and commercial vehicle outlets in Pinetown
27. Sandown Motors Holdings (Pty) Limited and Imperial Holdings (Pty)
Limited An agreement is to be entered into between the parties in terms of
which SMH will acquire from Imperial Holdings Ltd certain retail motor
outlets, being Mitsubishi Motors Cape Town, Cargo Northcliff and Cargo
Rosebank.
28. Imperial Holdings (Pty) Limited and Sirius Motor Corporation (Pty) Limited
The parties are to enter into an agreement in terms of which Sirius is
selling the Mercedes Benz franchise rights for Springs to Imperial, as well
as selling the Mitsubishi franchise rights for Gauteng East to Imperial. In
addition it is selling to Imperial the freehold property in Springs from where
the Union Motors dealership operates.
Extent of the mergers effect on the DCSA distibution network
29. The mergers affect only the five socalled metro centres and DCSA’s
29. The mergers affect only the five socalled metro centres and DCSA’s
distribution network outside of this remains unchanged. The reason for
this is that dealers outside these areas have a much lower turnover in
DCSA vehicles and it would not make sense for them to establish
separate brand show rooms for each of the DCSA‘s brands. Their
distribution network is thus considered an exception to the DNS strategy.
30. The tables below reflect the outlets which distribute DCSA‘s products in
the five metro centres and how the ownership of each will change post
6
merger. A distinction is made between commercial vehicles and
passenger vehicles. 9
COMMERCIAL VEHICLES
31. Commercial vehicles are typically classified according to light commercial
vehicles (“LCVs”), medium commercial vehicles (“MCVs”), heavy
commercial vehicles (“HCVs”) and buses & coaches.
Table 1: Gauteng North 10
Dealership Owner –Pre Merger OwnerPost
Merger
McCarthy Freightliner, Pretoria McCarthy SMH
McCarthy Truck Centre,
Centurion11
McCarthy SMH
Table 2: Gauteng East
Dealership Owner Premerger Owner Postmerger
Sandown Truck Centre, Kelvin SMH To be relocated to
another brand centre
Cargo Wadeville, Wadeville Imperial Imperial
Table 3: Gauteng West
Dealership Owner –Pre merger Owner –Post
merger
Garden City Roodepoort (CV) Barloworld SMH
9 Note that this delineation between the two is not meant to constitute them as relevant markets
for the purpose of the competition assessment. The relevant markets are more likely to be sub
groups of both these two broad categories. It is however convenient to organise the analysis in
this way as there are different considerations that arise in the respective commercial and
passenger sectors.
10 These outlets were taken from maps submitted by parties in their latest submission, under tab
A16. Ref:page 68990, 790 [Competitivness Report] –all outlets selling anything other than
commercial vehicles alone are grouped under PC’s. LCV’s occupy a category between CV’s and
PC’s. Therefore in analysing the passenger vehicle market, we have included cars that would
theoretically be classified as LCV’s . Italics denote change in ownership.
11 Denoted as separate outlet to McCarthy Freightliner on map under tab A6
7
Table 4: Durban
Dealership Owner –Pre
merger
Owner –Post
merger
McCarthy Truck Centre Pinetown
(CV)
McCarthy BML & Durban
South J.V
NMI Prospection Barloworld
Table 5: Cape Town
Dealership Owner Owner –Post
merger
McCarthy Truck Centre, Montague
Gardens
McCarthy SMH
Sandown CV Bellville SMH SMH
32. What we observe from the above is that in at least two metro centres the
number of different dealer outlets goes from two to one. In at least two
centres the number of dealers remains the same, although the identity of
the dealer may have changed. The parties argue that in respect of
commercial vehicles, the market is national and the alteration of
ownership in the metro centres, even if it leads to fewer players in certain
centres, is irrelevant, as the number of players nationally remains
unaltered. The reason they argue that the market is national, is because
the customer profile is different from that in the passenger vehicle market.
The typical commercial vehicles customer would be a firm buying several
vehicles for its fleet and which is not inconvenienced by sourcing from
anywhere in the country. Given that most commercial vehicles are
considerably more expensive on average than passenger vehicles,
customers are not reluctant to spend the extra time in travelling to source
the best deal.
33. We accept this argument and we have no reason to believe that the
restructuring insofar it affects commercial vehicle outlets of DCSA will lead
to any meaningful diminution of competition.
34. There remains as well significant interbrand competition in the markets
for MCV’s and coaches and busses. We set these tables out below.
Table 6: Market Shares of MCV’s 12 20012002
12 Commercial vehicles with a weight of between 7.5 and 16 tons.
8
Firm Market Share
Mercedes Benz 24.6%
Nissan 25.2%
Toyota 22.2%
Isuzu 15.6%
M.A.N 8.8%
Iveco 2.3%
Freightliner 0.9%
Volvo 0.45
ERF 0.1%
Source: MB Commercial Vehicles Business Plan 2001/2
Table 7: Market Shares of HCV’s 13– 20012002
Firm Market Share HCV’s
Mercedes Benz 21.6%
M.A.N 18.6%
Navistar 14.1%
Scania 9.8%
Freightliner 9.1%
Volvo 6.3%
Nissan 5.6%
Peterbilt 4.8%
Toyota 3.3%
Iveco 2.4%
Isuzu 1.7%
ERF 1.5%
Mack 1%
Source: MB Commercial Vehicles Business Plan 2001/2
Table 8: Market Shares of Buses– 20012002
Firm Market Share HCV’s
Mercedes Benz 35.5%
M.A.N 28.4%
Iveco 15.8%
Volvo 9.4%
ERF 6.3%
Scania 4.6%
Source: MB Commercial Vehicles Business Plan 2001/2
13 Commercial vehicles with a weight over 16 tons.
9
35. The possible exception may be the category of light commercial
vehicles14. This category is an amalgam of light delivery vehicles and
passenger utility vehicles or fourwheel drives. Again, it is not necessary
for us to delineate as the relevant product market, if it be one, any further,
as there is sufficient interbrand competition for both LCV’s and fourwheel
drives to alleviate any concerns. Below for the sake of completeness we
set out market shares in respect of the sale of light delivery vehicles.
Table 9: Market shares Light Commercial Vehicles, July 2002
Firm Market Share
Daimler Chrysler 9.07%
Delta 18.56%
Fiat .59%
FMCSA 20.79%
Nissan 18.75%
Renault .51%
Toyota 28.26%
Volkswagen 3.47%
Source: MB Commercial Vehicles Business Plan 2001/2
36. Although the market is moderately concentrated and DCSA has a share of
9.07%, there are a number of players in this market who have significant
shares.
37. We conclude that in the commercial vehicle market viewed as a whole, or
by segment, there will be no loss of either intrabrand or interbrand
competition as a result of the transactions. As far as intrabrand
competition is concerned the reconfiguration of the metro areas has made
little difference and to the extent that it may in certain categories which
may be more local than national, interbrand competition remains
sufficiently vibrant across all categories.
38. We will therefore focus the remaining part of our analysis on the
categories within the passenger vehicle market .
PASSENGER VEHICLES
39. We set out in the tables below the change in the number of players in
each of the sector by comparing the premerger and post–merger
14 Described as a vehicle with a weight of 3.5 to 7.5 tons.
10
positions.
Table 10: Gauteng North
Dealership Owner Pre
Merger
Owner Post Merger
McCarthy Fountains, Pretoria McCarthy SMH
Ellenby Motors, Hatfield SMH McCarthy
McCarthy Menlyn McCarthy McCarthy
McCarthy Centurion Park, Centurion McCarthy McCarthy
Mitsubishi Centurion SMH McCarthy
Grand Central Motors, Midrand Independent Independent
Table 11: Gauteng East
Dealership Owner Owner Post Merger
Mercurius Motors, Kempton Park 15 Imperial Imperial
Cargo Edenvale Imperial Imperial
Cargo M2 City, Johannesburg Imperial Imperial
Mitsubishi East Rand, Boksburg Sirius Motors Imperial
Mercurius East Rand, Boksburg Imperial Imperial
Cargo Germiston, Germiston Imperial Imperial
Mercurius Alberton, Alberton Imperial Imperial
Union Motors Springs 16 Sirius Motors Imperial
Table 12: Gauteng West
Dealership Owner Owner Post Merger
Garden City Krugersdorp 17 Barloworld SMH
Mitsubishi Sandton SMH SMH
McCarthy Randburg McCarthy SMH
Garden City, Roodepoort Barloworld SMH
SMH Sandton SMH SMH
Cargo Northcliff Imperial SMH
Cargo Rosebank Imperial SMH
Cargo Auckland Park Imperial Imperial
Shiraz Auto Lenasia Independent Independent
Century Motors Carltonville Independent Independent
15 Sell new CV’s and PC’s
16 Sell new CV’s and PC’s
17 Service and Parts outlet only
11
Table 13: Durban
Dealership Owner Pre
Merger
Owner Post
Merger
McCarthy Pinetown (PC) McCarthy BML & Durban
South J.V.
Durban South Motors, Winklespruit Independent BML & Durban
South J.V.
Mitsubishi Umhlanga Rocks Independent BML & Durban
South J.V.
NMI Umhlanga Barloworld BML & Durban
South J.V.
NMI Durban Barloworld BML & Durban
South J.V.
Table 14: Cape Town
Dealership Owner Pre
Merger
Owner Post
Merger
Malmesbury Motors, Malmesbury Independent Independent
McCarthy Milnerton McCarthy SMH
Orbit N1 City SMH SMH
Paarl Motors, Paarl Independent Independent
Eikestad Motors, Stellenbosh SMH SMH
Mitsubishi Motors, Paarden Eiland IHL SMH
McCarthy Culemborg McCarthy SMH
McCarthy Claremont McCarthy SMH
Rola Motors, Strand Independent Independent
40. A comparison of the pre and post merger situations reveals exactly what
the parties say it does – there will be fewer owners in the metro areas.
What is not apparent from this comparison, but the parties concede this, is
that these fewer owners will enjoy their franchise over a larger area than
they enjoyed previously. This may be slightly misleading. All that this
means is that DCSA will not grant another franchise within that metro
centre to another dealer. It does not preclude a franchisee from dealing
with customers who are situated in another metro. 18 We are thus not
dealing with a situation where there are de jure exclusive territory
allocations but they may nevertheless, and this we discuss further below,
18 At the hearing, the parties pointed out that the territorial demarcations will be maintained but
the dealers will be entitled to advertise nationally. ( See Transcript pg 68 )
12
operate post merger as a de facto exclusive allocation of territories.
Competitive Analysis of the passenger vehicle market
41. We have previously stated that a reduction in intrabrand competition is
only of concern if there is weak interbrand competition. 19 For this reason
we will approach our decision in two stages. First, we will examine
whether there has been a reduction in intrabrand competition. Only if we
find that there has do we need to consider the state of interbrand
competition in order to determine if the merger raises competition
concerns.
Intrabrand competition in the passenger vehicle markets
42. The RMI raised various concerns with regard to the Dealer Network
Strategy. In relation to interbrand competition, in their written submission,
they suggested that the merger be approved subject to two conditions to
protect intrabrand competition. Firstly, that DCSA may not prevent
dealers from conducting sales via there own websites and secondly that
there be no prohibition on franchised dealers carrying out sales to
independent resellers. It is not clear whether the new DCSA franchise
agreement, which we are advised is still in the process of being
negotiated, will contain these restrictions. We will not consider the merit of
these proposals now. Later in this decision we set out the approach we
have taken in relation to all the conditions that the RMI have suggested.
43. The extent of intrabrand competition in the metro centres premerger is
difficult to determine. The Tribunal was given mixed messages by the
parties throughout the documentation and during the hearing. The DNS
documentation reveals a great preoccupation with preventing the dealers
from competing with themselves as did evidence at the hearing:
from competing with themselves as did evidence at the hearing:
44. According to the testimony of DCSA’s Mr van Olst dealers at present were
too focussed on intra rather than interbrand competition.
“ Certainly, it is our contention that dealers focus on the intrabrand
competition rather than on the interbrand. And that is the reality on
what is happening. It is to get the deal to come through my door,
rather than some other dealer of the same brand’s door, rather than
try and compete with the BMW’s and the like” 20
19 DaimlerChrysler SA (Pty) Ltd/Sandown Motor Holdings (Pty) Ltd and
Nkosinauth Ronald Msomi & Others v British American Tobacco 49/IR/Jul02, at para 49.
20 See Transcript page 26.
13
45. Similar sentiments were echoed by Mr.Kopke the chief executive officer.
The documentation submitted by the parties detailing the DNS strategy
also shows DCSA’s preoccupation with the perceived competition
amongst its dealer network. A sample of the cryptic sound bites of their
presentation package illustrates this:
“DCSA franchise is overrepresented ,” or “profiteering at whatever
cost”, or “interdealer rivalry rather than the competition”, or
“Dealerships need to avoid intranetwork competition, or “Minimise
and hopefully eliminate the negative effects of current discount
practice..” 21
46. In another document one of the weaknesses identified is “ inter –dealer
competition”
47. On the other hand, the dealer perception of this level of competition was
more muted. Mr Michaux from Cargo Motors stated in his testimony that
competition is based on customer retention:
“Sure there is intrabrand competition that takes place, but it really
is not I think of the magnitude that the discussion talked earlier
on… our closeness to our customer and the opportunity to sell a
customer and have a customer for life, for us, now I am talking as
Cargo Motors as a dealer versus the other dealers. It is a big issue
for us. We pride ourselves in customer retention.” 22
48. One of the difficulties in assessing the level of intrabrand competition is
that pricing is not transparent. As one witness put it, price competition
takes many forms and includes the provision of extras and trading
allowances.23
49. The stated intention of the new strategy is to have fewer sales points and
more service points to conveniently serve customer’s servicing
requirements. There is little doubt that it is DCSA’s stated intention to
reduce whatever level of intrabrand competition already exists in the five
reduce whatever level of intrabrand competition already exists in the five
metro areas. The new structure is likely to inhibit intrabrand competition
as dealers are situated further apart, physically leaving only certain
customers, located on the peripheries of two overlapping metro areas,
with the ability to shop around without having to travel too far.
21 DCSA Strategy Workshop Document July 2000
22 Transcript page 56.
23 Transcript page 47.
14
50. Two other factors will further inhibit dealers. First, dealers are required to
make substantial investments in the new showrooms envisaged in terms
of the DNS, estimated by the parties to be in the order of some forty
million Rand. They are therefore unlikely to want to risk that investment by
angering DCSA, particularly if the franchise can be cancelled at short
notice. Second, as DCSA increasingly integrates further into the
downstream market, through its investment in Sandown Motors, its ability
to monitor the behaviour of other dealers is increased, whilst at the same
time its incentives to reduce intrabrand rivalry also increase, as it now
competes with its own customers in distribution.
51. We have come to the conclusion that the combined effect of a physical
network, in which there will be fewer dealers selling in the relevant
geographical areas, and the increased disincentives faced by dealers to
compete inter se, will lead to a reduction in intrabrand competition.
Whether the reduction is sufficient to meet the test of ‘substantiality’
required under section 12 A, is more difficult to answer, as we do not know
to what extent there was intrabrand competition in these areas before the
merger. However, we do not need to be more precise about this aspect on
account of our conclusions about the level of interbrand competition.
Interbrand competition in the passenger vehicle market
52. Having concluded that there is, in all likelihood, going to be a reduction in
intrabrand competition, we need to examine whether interbrand
competition will be substantially lessened.
53. There have been profound changes to the motor vehicle industry since
1994, with the advent of our new constitutional dispensation. The political
changes meant that the industry ended its years of isolation. Tariff barriers
changes meant that the industry ended its years of isolation. Tariff barriers
dropped, imports entered the market and a concerted industrial policy to
streamline and revitalise the sector transformed the competitiveness of
domestic manufacturing.
54. It is beyond the scope of this decision to analyse how and why these
changes have come about. What is important is what they have brought
about, a market in which we have a large number of manufacturers
actively competing against one another, many with a large portfolio of
brands across a wide segment of the market.
55. The presence of many brands of motor vehicle was confirmed at the
hearing by Mr Michaux:
15
“[In the] South African market we have almost every
single vehicle available for sale that is available
internationally… I mean, if you look at vehicles like SAAB,
Lexus, the new Peugeot that has come out, Bentley I
believe is coming now. We really compete with
everybody.”24
56. Mr Köpke remarked that in contrast to 1994, when there were
approximately 6 manufacturers in the country, there are now a
plethora of car brands on the market:
“Certainly the exclusivity strategy that we are proposing has
certainly not been a huge hindering factor for Peugeot, for
Jaguar, for all those brands that have come in. They have come
in very quickly. They have come in with very good dealers.
They have had good backing and certainly they have all
become competitors of ours, so the exclusivity strategy that we
have is certainly not inhibiting new players in any way coming
into this market place.” 25
57. The prevalence of many different brands in each market segment is
testimony to the ease of entry of continually new manufacturers into the
market.
58. The tables below reflect the proliferation of brands in all the relevant
markets. Since the dealer network strategy affects DCSA franchises, the
relevant segments in which DCSA is active are small cars, luxury cars,
speciality cars, utility vehicles and minivans. 26
59. The tables below reflect, where available, the regional and national market
shares per segment and the significant DCSA competitors in all market
segments in which it competes:
24 Transcript pg 59.
25 Transcript pg 91.
26 In March 2002 the Mercedes Benz Smart car entered the SA market under the Mini Car
segment. These market shares are insignificant, therefore not reflected. Although we reflect
market shares in terms of these narrower categories, we accept, as we did in the last merger, the
parties’ arguments that there is much demand substitutability between upper and lower luxury
cars, as well as between lower and upper speciality vehicles.
16
Table 15: Market Shares according to Segment– 2002
Segment Brand
Gauteng m/s Western
Province m/s
KwaZulu Natal
m/s
Total National
Market
Share27
Small Cars Mercedes
Benz
9 2 7 4.82
Chrysler
Neon 2.96
VW 29 14 34 29.43
Toyota 15 51 19 19.31
Mazda 14 12 12 16.91
Other
competitors Nissan, Honda, Alfa, Audi, Daewoo, Ford, Opel, Peugeot, Renault, Subaru
Lower
Luxury
Mercedes
Benz
30 37 44 37.05
BMW 44 30 36 39.57
VW 19 20 15 2.14
Other
competitors Alfa, Audi, Toyota Lexus, Opel, Saab, Volvo
Upper
Luxury
Mercedes
Benz
21 55 34 27.87
BMW 48 23 42 46.28
Other Alfa, Audi, Cadillac, Jaguar, Toyota Lexus, Peugeot, SAAB, Volvo
Lower
Speciality
Mercedes
Benz
34 6 47 36.79
BMW 52 69 36 46.98
Other Mazda, Opel, Peugeot, Renault,
Upper
Speciality
Mercedes
Benz
29 37 38 31.03
BMW 37 44 43 40.63
Other
competitors Alfa, Audi, Honda, Jaguar, Porsche, SAAB, Toyota, Volvo
27 According to DCSA Market Analysis July 2002
17
Small Utility Mitsubishi
Pajero 1.42
Chrysler
Jeep
Wrangler
3.16
Landrover
Freelander 21.18
Toyota 48 62 54 58.49
Other Opel, Renault, Suzuki,
Lower
Middle Utility
Jeep
Cherokee
56 52 46 51.32
Isuzu 32.76
Landrover 8.84
Nissan 21 27 28 24.79
Other Honda, Isuzu, Subaru,
Upper
Middle Utility
Jeep Grand
Cherokee 19.04
Mercedes 15.58
Mitsubishi
Pajero 14.04
BMW 38 38 35 20.39
Landrover 16.26
Toyota 5 21 5 6.54
Other Audi, Chevrolet, Nissan, Volvo
Small
Minivans
Chrysler PT
Cruiser 17.65
Renault 78.22
Daewoo 4.12
Minivans
Chrysler
Grand
Voyager
66.73
VW 30.47
Peugeot 1.4
Renault 1.4
18
Source: DCSA Market Analysis July 2002
60. As the parties pointed out at the hearing, in the high end of the
market, that is, luxury vehicles, sports utility vehicles and
specialised vehicles, there is now a large range of imported
brands that are in direct competition with the Daimler Chrysler
brands. Their convincing levels of market penetration suggests
that they have had no difficulty whatsoever in finding
distribution facilities within the country and are able to compete
vigorously.
61. In addition, there have been major swings in market share within each
segment in recent years. We have not wished to burden this decision with
all the tables we have received evidencing this, but it will suffice to give a
few examples. For instance;
in the lower middle utility sector, Isuzu went from a market share of
32.22% in 1998 to 30% in 2001 to 4% in 2002. Land Rover went from
52% in 1997 to 20% in 2002 to 9% in 2002.
In the lower luxury segment, Audi went from 20% market share in 1997 to
14.8% in 2002.
In the upper luxury segment, Mercedes went from 48% in 1997 to 27.87%
in 2002.
62. This erosion of market shares suggests that interbrand competition is
alive and robust within each market segment.
63. There is no doubt that BMW is a major competitor, dominating the luxury
and speciality segments and, along with Toyota in the utilities segments,
Chrysler is dominant in the minivan segment. However in this and all the
other segments there are a significant number of other players in the
market to constitute sufficient competition.
64. We have thus far approached the mergers by examining the likely
negative effect on intrabrand competition. However as the literature on
the subject reflects, a reduction in intrabrand competition is frequently
the subject reflects, a reduction in intrabrand competition is frequently
accompanied by an increase in interbrand competition. In the case of
these mergers we have no reason to doubt that the same will occur.
Balanced against the probable loss of intrabrand competition the mergers
will make the realigned DCSA metro network a more formidable
competitor to its rivals like BMW and Audi.
19
65. As the brand centre concept will entail dedicated marketing and resource
focus on specific brands, the chances that such a strategy will enhance
interbrand competition between manufacturers of different brands is
relatively high. This is reinforced by the fact that brand value in motor
vehicles seems, more than in any other market, to command great
power.28 We therefore conclude that interbrand competition is and is
likely to remain vigorous and there will be no substantial lessening of
competition in any of the relevant markets.
66. This does not conclude our analysis however as the dealer networks do
not simply sell cars to retail buyers, they also offer service and sell spare
parts. We consider these two issues below.
Servicing
67. The parties were questioned at the hearing as to what extent consumers
would have a choice of dealer if they were unhappy with the service or
price of one service dealer, particularly since there will be one dealer
group per metro area. There was a concern that there would be a reduced
choice in service outlets. Though there might well be a reduction in the
choice of service outlets per area, this is mitigated by the fact that the
price of such service is known upfront, since it is incorporated into the
maintenance plan sold with the vehicle. Though there are slight
differences between dealers in hourly labour rates, generally the
maintenance plans mean that the service and repair cost is factored into
the total purchase price. Therefore the consumer will tend to consider the
pricing of the service as part of his or her overall purchase decision. The
parties argued that as far as service quality is concerned, there is no
obligation on consumers to go to an authorised dealer. Therefore if service
obligation on consumers to go to an authorised dealer. Therefore if service
levels deteriorate, he or she would be able to go elsewhere to service their
vehicles. Some customers who are located conveniently in relation to
more than one metro area will still have the choice of more than one
outlet.
68. There of course is the risk that use of a nonauthorised dealer may lead to
service by people not competent in service of the vehicle or use of inferior
parts and thus for many customers may not be an attractive alternative.
On the other hand it is also true that specialist service outlets are now
providing a growing number of services to the motorist, such as batteries,
tyres and exhausts.
28 Mr Kopke confirmed that the South African market is unique in its degree of sophistication with
respect to motor vehicle retailing.
20
69. Thus the class of consumer who may be faced with high service prices
and who cannot substitute these services without great inconvenience
may be relatively small , once we exclude those who have a maintenance
contract, those who can get the service performed by another DCSA
dealer or require a service that is not by its nature required to be
performed by a DCSA service specialist .
Spare Parts
70. Like in the market for servicing, the market for spare parts is more local
since customers typically do not wish to travel too far to obtain parts.
Parts are bought for the maintenance and servicing of motor vehicles.
Dealers generally source parts from the manufacturers of the brands they
sell. Therefore they will stock only those parts of the particular DCSA
brand they sell. According to the parties, 35% of DCSA parts are used in
the dealers’ own workshops. The other 65% are sold to retail customers,
nonfranchised dealers such as fleets, the repair industry and independent
workshops.
71. In respect of pricing of spare parts, the parties in their papers do suggest
that spare parts are invoiced to dealers at a premium margin for the
manufacturer. Discounts are offered to customers, dependant on their
size. Therefore individual consumers may actually receive very little
discount on the cost of spare parts. However, the parties do maintain that
there is presently nothing to prevent dealers from selling spare parts to
independent repairers and the new DNS will not alter this, leaving the
competitive nature of the spare parts market unchanged. Furthermore it is
stated that independent companies compete vigorously with DCSA
dealers in that they sell nongenuine, equivalent spare parts which
dealers in that they sell nongenuine, equivalent spare parts which
account for about 40% of all spare parts demand for DCSA product. There
is no evidence that the new strategy would diminish this alternate source
of parts. In any event, the same considerations apply in respect of
maintenance plans as do for servicing. The cost of spare parts will be
factored into the customer’s ultimate purchasing decision 29.
72. The brandspecific outlets will also offer servicing and parts sales in
respect of that particular brand. The parties assert that not the number but
merely the positioning of dealers selling the parts will change and there
are likely to be more parts and servicing outlets. The DNS strategy will not
change the marketing and sales of spare parts at all.
73. Nevertheless our concerns about the ability of dealers to exercise market
29 Provided the customer utilises original spare parts.
21
power in relation to parts and service, post merger, were the ones least
satisfactorily dealt with by the parties, but they are insufficient to change
our conclusion reached earlier that the merger will not pose a competitive
concern because of strong interbrand competition.
74. The evidence suggests that consumers in this market are sophisticated. If
the DCSA dealer network wishes to raise service prices beyond a
competitive level the consumer is likely, at least in the medium term, to
become aware of this and to favour the products of its rivals. This is likely
to inhibit the exercise of market power by the dealers in the metro areas in
relation to the cost of services.
Conditions
75. We referred above to the fact that the RMI had argued that the
merger be approved subject to certain conditions. The two
conditions that related to the loss of intrabrand competition
were referred to earlier in our analysis. 30
76. Two further conditions were suggested that relate to both intrabrand and
interbrand competition. Both conditions would require us to impose
conditions on what DCSA’s agreements with its franchisees may contain.
We pointed out earlier that the new franchise agreements have not yet
been concluded. We know what the present agreements contain, but the
parties inform us there is an ongoing process of negotiation in respect of
the new ones. We are certainly in no position to secondguess the
respective hands of the negotiators at the bargaining table. The
manufacturer’s leverage is clear, but the dealers involved in this merger
are all large concerns who DCSA can ill afford to alienate. Nevertheless
the absence of an agreement is not in itself a bar to the imposition of
conditions if they are appropriate. We will go on to consider their
appropriateness.
77. The first condition was:
appropriateness.
77. The first condition was:
“that there shall be no direct or indirect noncompete obligation
relating to the sale of motor vehicles and that multibranding by any
30 See paragraph 42
22
dealer shall be possible as set out paragraph 4.2.1…”
78. It needs to be pointed out that presently there is no prohibition
on a firm of dealers from selling the products of its rivals.
Imperial, McCarthy and Barlows all retail the vehicles of a
number of rival manufacturers. What most manufacturers
prohibit their franchisees from doing is selling the rival product
from the same showroom. Hence we have branches of the
same firm of dealers with different product brandings, McCarthy
Audi, McCarthy BMW etc.
79. The mergers do not alter the status quo in this regard. The RMI
seem to be wanting to utilise the mergers as an opportunity for
us to impose a structural change to the industry that we have
never had before in this country. It is thus not a condition, which
would restore the status quo, but one that seeks to restructure
the industry into something it has never been. We do not seek
to dismiss the RMI’s major thesis, which is that consumers
would be better served by having showroom floors which stock
a range of rival manufacturers offerings. Retail outlets on this
model would become giant supermarkets selling a range of
brands. This, they inform us, is the direction that the European
Union would like to see the market moving with its new block
exemption. 31
80. We must consider if a merger is likely to substantially lessen or prevent
competition and only if the answer to that is, ‘yes’, after the public interest
and efficiency balance have been taken into account, may we consider
prohibition or impose conditions. Without that prior conclusion we have no
jurisdiction to prohibit a merger or impose conditions.
81. It might well be, and we express no view on this, that another model of
distribution of motor vehicles will increase rivalry and lead to lower prices.
31 According to the new block exemption, July 2002, dealers can choose between selective or
31 According to the new block exemption, July 2002, dealers can choose between selective or
exclusive distribution. They will be allowed to sell competing motor vehicle models off the same
showroom floor. If manufacturers allocate exclusive sales territories to dealers, they cannot stop
them from selling to independent resellers. Furthermore, if dealers are not authorised to sell to
independent resellers (selective distribution), that is supermarkets, internet resellers, then
manufacturers will no longer be allowed to stop their dealers from setting up dealerships in other
selective territories than the one in which they are authorised.
23
It is not our function, however, to use merger control to either reregulate
or regulate an industry’s structure.
82. The same can be said for the remaining condition which states:
“That, to prevent a manufacturer from terminating a franchise
agreement because a dealer engages in procompetitive behaviour
–
5.2.3.1 every notice of termination must clearly state the reasons
for the termination;
5.2.3.2 one years notice of termination has to be given if a network is
reorganised or if compensation is paid to the dealer, and two years’ notice has to
be given in all other cases; and
5.2.3.3 in the case of termination of a contract but also where disputes arise
regarding contractual obligations, the parties shall refer such disputes to an
independent expert third party or arbitrator.”
83. It is certainly true a manufacturer could use a condition that allows it to
terminate dealers at short duration, to discipline a maverick, but pro
competitive dealer. But it does not follow that by imposing a condition
obliging DCSA to give longer notice periods to terminate their dealer
contracts that this would inevitably have the effect of enhancing
competition between the dealers. It is equally arguable that it could lead to
the entrenchment of inefficient dealers.
84. The parties maintained that imposing either of these conditions would
entail a selective intervention, in that it would affect only DCSA franchise
contracts, and not those of other manufacturers. Though there might be
some cases where such a selective intervention is justified, we do not find
there is just cause in these merger proceedings. For all the above
reasons, and in the absence of anticompetitive mergerspecific effects,
we do not deem it appropriate to impose any of the conditions on the back
of a merger enquiry.
of a merger enquiry.
For all the above reasons we approved this series of mergers unconditionally.
There are no public interest concerns, which would alter this conclusion.
24
_____________ 19 November 2002
N. Manoim Date
Concurring: D. H. Lewis, F. Fourie
For the parties: Adv D. Unterhalter instructed by Deneys Reitz Attorneys
For the Commission: Mr A. Coetzee and Ms. L. Blignaut
25