COMPETITION TRIBUNAL
REPUBLIC OF SOUTH AFRICA
Case Number: 06/LM/Oct99
In the large merger between
Highveld Steel and Vanadium Corporation Ltd
Van Leer South Africa
and
Steelbank Merchants (Pty) Ltd t/a Drumpak
________________________________________________________________
Reasons for Competition Tribunal’s Decision
________________________________________________________________
Approval
The Competition Tribunal approves the merger without conditions. The reasons
for our decision are set out below.
The Transaction
1. The transaction under consideration is the acquisition of the assets of
Drumpak, a division of Steelbank Merchants (‘Steelbank’), by Rheem
South Africa (‘Rheem’), a division of Highveld Steel and Vanadium
Corporation Limited, a subsidiary of the Anglo American Corporation of
South Africa Limited, and by Van Leer South Africa (Pty) Limited (‘Van
Leer’), a wholly owned subsidiary of Royal Packaging Industries Van Leer
NV, a company incorporated in the Netherlands. The parties to the
transaction all manufacture steel drums that are principally used for
transporting and storing products of the oil, chemical and food industries.
2. The transaction originated in an approach by Steelbank to Rheem. Rheem
initially rejected a proposal that it purchase the assets of Drumpak
whereupon Drumpak approached Van Leer. Van Leer avers that it
elected to purchase the assets of Drumpak because it did not want them
to fall into the hands of its competitor, Rheem, and because certain of the
assets in question could be profitably applied in conjunction with Van
Leers existing assets. Van Leer and Rheem ultimately agreed to jointly
purchase the assets of Drumpak.
3. The assets in question comprise certain plant and machinery and the
intellectual property rights of Drumpak. The total consideration is R 9 642
525. According to the sale agreement between the parties, the
consideration was structured to reflect a value on the material assets of
R781 486, a further R2 000 000 in respect of intellectual property and R 1
461 039 for goodwill in respect of a restraint by Steelbank. In addition, the
agreement provides for the payment of a further R5 400 000 in exchange
for noncompete agreements with four executives in respect of the
manufacture of a variety of industrial container products. However, in the
course of hearings before the Tribunal, the parties, in response to queries
regarding the striking premium paid on asset value, insisted that their
valuation of the assets was, in reality, considerably higher than that
reflected in the formal structure of the deal. We will return to this point
below.
4. The acquiring firms then entered into a separate side agreement between
themselves. They agreed that Van Leer will export certain of the assets to
Nigeria and will present proof of this to Rheem. Rheem and Van Leer will
reach agreement with respect to the division of the remaining assets
between them, failing which they will be scrapped. The assets that will be
shipped to Nigeria include a welder and corrugator, key pieces of
equipment in the manufacture of new steel drums.
5. Steelbank avers that it elected to sell the assets because of the depressed
state of the steel drum market and growing competition from the increased
availability of substitute products. Its profitability is further threatened by
the removal of a decentralisation grant that effectively enabled it to
subsidise the price of its product and, hence, its market share.
subsidise the price of its product and, hence, its market share.
The Relevant Market
6. The parties have defined the relevant product market as the industrial
containers market which includes bulk containers, plastic drums,
reconditioned drums, bags, bulk tanks, bulk road tanks and fibre drums.
The parties aver that, in the main, these are substitutes although they
acknowledge that there are minor circumstances where the products in
the relevant market may not be substitutable.
7. The Commission contends that the relevant market is somewhere
between new steel drums and industrial containers, that is, the
Commission acknowledges a degree of substitutability between new steel
2
drums and other industrial containers. Accordingly the availability of
containers other than new steel drums will play a role in disciplining the
pricing decisions of producers of new steel drums, but to a rather limited
extent. However, the Commission has not been able to provide evidence
that would specify more precisely the degree of substitutability between
new steel drums and other industrial containers, and, therefore, the extent
to which the availability of the latter disciplines the prices of the former. In
fact, at the hearing of this matter, counsel for the Commission informed
the Tribunal that the question whether ‘arbitrage’ would take place
between new steel drums and the proposed alternatives had not been
investigated by the Commission. In addition, it is far from clear where the
Commission considers the boundaries of the market for industrial
containers to be. Its calculations of market concentration in Annexure I to
its recommendations imply a rather limited market for industrial containers,
comprising only plastic and steel containers (Scenario 4 in Annexure I),
while section 2 of the report contains a more extensive list of possible
substitutes, which suggests a rather wider market for industrial containers.
8. The Commission further holds that the geographic market is confined to
KwazuluNatal. The merging parties did not seriously take issue with the
Commission’s definition of the geographic market. Although Rheem
criticised the Commission’s conclusion in this regard in its written
submissions to us, it failed to put forward a convincing argument for a
wider geographic market. We therefore accept the Commission’s
recommendations on the geographic market.
9. In essence, the merging parties’ interpretation of the evidence is that the
9. In essence, the merging parties’ interpretation of the evidence is that the
opportunities for substituting alternative products for steel drums are
considerable thus supporting a market definition that incorporates a wide
range of alternate products. Indeed the target company attributes the
decline, since 1994, in the volume of sales of steel drums, ‘...largely to an
increase in the demand for and availability of substitute products.’ In
support of this claim the various parties argue that
• Technological developments in the plastic industry have
improved capacities, manufacturing costs and product
performance resulting in a competitive product;
• 1000 litre plastic intermediate bulk containers are the fastest
growing industrial packaging container worldwide;
• Reconditioned drums are substitutes for new steel drums. It
is pointed out that 4 200 chemical drums enter South Africa
each and every day and that these are reconditionable;
• Actual substitution is already occurring. A plastic drum
3
manufacturer is tendering for contracts from large oil
companies and that most of the oil industry is already using
1000L IBCs and reconditioned drums.
10. The parties do however acknowledge, without further elaboration, that
each product will have its own advantages and disadvantages that will
equate to a value which would be different to each customer.
11. The Commission’s evidence, based on interviews with 24 of the 63 largest
customers identified by the parties, conflicts sharply with that offered by
the parties. Most pertinently, these customers point out that
• Toxic, highhazard and flammable products can only be transported
in new, 210 litre steel drums;
• Products including solvents, oil and chemicals can only be shipped
and stored in new steel drums;
• Substitutability in the food industry is limited because a large range
of food products are susceptible to contamination when stored in
plastic containers;
• Plastic drums are considerably more expensive than steel drums
and that they would not switch to plastic drums in response to a
510% increase in the price of steel drums;
• They would not readily switch to reconditioned drums because
these products are not reliable or guaranteed and because of the
poor image that they convey for the company;
• It is not economically viable to transport empty drums from
suppliers in Gauteng and that accordingly they only purchase
drums from suppliers in KwazuluNatal.
12. Drumpak challenged the validity of the Commission’s representations
based on interviews with the parties’ customers, alleging that they were
inaccurate and biased for a number of reasons. These included that –
• Drumpak had obtained written confirmation from customers
contradicting some of the Commission’s conclusions – for instance
confirmation that other containers are substitutes for new steel drums
confirmation that other containers are substitutes for new steel drums
for purposes of storing and transporting oil;
• many of the customers interviewed were insignificant players or did not
buy steel drums in Kwazulu Natal.
13. Neither party has presented compelling evidence in support of their
conflicting assertions on the degree of substitutability. While the Tribunal
considered using its inquisitorial and other powers to generate more
4
conclusive evidence this would have caused a significant delay in the
already prolonged evaluation of this transaction, a delay that has already
caused considerable prejudice to a small party.
14. There are inferences that support either conclusion. On the one hand, the
structure of the purchase price suggests that the parties themselves
implicitly accept a narrower market definition than that asserted. A number
of factors in the way the deal is structured have contributed to our
scepticism:
• Although the parties conceded that the difference between
the total contract price and the material assets was greatly
inflated, they were unable to put a credible value to the
assets, but assured us that the sale agreement considerably
undervalues the assets. Yet the side agreement between the
purchasers in which they inter alia divide the assets inter se
seems to value the assets consistently with the valuation
placed in the sales agreement when one would have
considered the discrepancy to have manifested itself at this
stage when the purchasers divide their spoils;
• Restraints on entering the market again are extensive and
bind not only Steelbank, but its key executives, which is
unusual in a transaction in a business not being sold as a
going concern;
• Steelbank is further restrained from using certain productive
capacity not part of the deal from competing in the container
market or being sold to anyone who will use it directly or
indirectly for the manufacture of drums;
• In their side agreement, the purchasers agreed to the
division of the assets on the condition that the most
productive assets were given to Van Leer to be used in the
Nigerian market as we have outlined above.
The parties suggested that the purchase consideration was structured in
The parties suggested that the purchase consideration was structured in
the way that it was to suit the sellers and that the value put on the assets
in that structure does not accord with their actual commercial value.
However, absent a true commercial valuation of the assets, the Tribunal is
unable to assess whether the premium is accounted for by the difference
between the stated asset value and the actual commercial value, or
whether the premium represents a payment for market domination thereby
reflecting the capitalized value on the price and market share premium
that they expect to glean in consequence of a diminution of competition.
Given that this premium for market domination would not be generated if
5
substitution occurred from producers of products alternate to steel drums,
it may be inferred that the parties themselves implicitly accept a narrower
market definition than that asserted.
15. Our skepticism of the transaction is further underlined by the resolution
passed by the directors of Highveld Steel on 18 August 1999, in which
they approved the transaction because it would give Rheem ‘the
opportunity … to participate in an additional market of approximately 200
000 new drums per annum’, which was the annual output of Drumpak
directly prior to the transaction. This clearly suggests that Highveld Steel’s
directors viewed the transaction as an opportunity to remove the capacity
of a competitor in the market for new steel drums. This would only be of
value to the purchasers if the relevant market was constituted by new
steel drums, that is, if new steel drums were not substitutable by other
products in the industrial container market.
16. On the other hand, given that the wellinformed and wellresourced
purchasers of steel drums have not approached the Commission or the
Tribunal, we may draw the inference that they believe that there are
substitutes in the market that will effectively discipline the price of new
steel drums.
17. In the event, the Tribunal is unable to make a conclusive finding on the
relevant market – in the unsatisfactory characterization offered by the
Commission, it falls ‘somewhere between new steel drums and industrial
containers’.
Market Power
18. Unfortunately, however, arriving at a firm conclusion regarding the post
acquisition share of market power depends, in significant part, on knowing
where on the spectrum ‘between new steel drums and industrial
containers’ the relevant market is to be found. We hasten to add that it is
containers’ the relevant market is to be found. We hasten to add that it is
near impossible to define relevant markets with absolute scientific
precision. However, the evidence offered by the Commission and,
alternatively, by the parties falls well short of the standard required for
conducting a market power analysis.
19. In assessing market power, that is in assessing whether the merger will
substantially prevent or lessen competition, the Act enumerates a number
of factors that are to be taken into account. These factors are ease of
entry into the market, the level, trends and history of collusion in the
market, the degree of countervailing power in the market, the dynamic
6
characteristics of the market, whether the business of a party to the
merger has failed or is likely to fail, and whether the merger will result in
the removal of an effective competitor.
20. The parties cite the activities of Sirco Systems as evidence of relative
ease of entry into the market for steel drums in Kwazulu Natal. Sirco is a
US based firm that produces new steel drums using a novel technology.
Essentially the component parts of the drum are imported and the drums
are built on the site of the customer. However, investigation by the
Commission reveals that Sirco has been selling steel drum kits in South
Africa for over two years in which time it has only been able to gain one
customer. In that period it has sold about 9000 drums to the customer
these being specialized resin drums that cannot be produced by any of the
steel drum manufacturers in South Africa. Most of the customers
interviewed by the Commission had not heard of Sirco, which does not
have an agent in South Africa. Moreover it appears that Van Leer’s parent
company owns a controlling interest in Sirco. Sirco does not offer
evidence of ease of entry into the market and, in any event, given its
relationship to Van Leer, is not an independent competitor.
21. While no evidence has been presented that suggests a previously
collusive relationship between Van Leer and Rheem, the process of this
transaction is a disturbing indicator of their ability to cooperate in the
removal of a competitor. This experience certainly enhances their ability
to collude and the postmerger structure of the market lends itself to
collusion. The Tribunal earnestly recommends that the Commission
closely monitor developments in this market.
closely monitor developments in this market.
22. Demand for new steel drums is concentrated amongst a relatively small
number of large companies in the oil, chemicals and food industries
suggesting a high degree of countervailing power. While it is not clear
how these admittedly powerful companies would exercise their power
against a duopoly of producers of new steel drums, their silence during the
course of this enquiry is conspicuous. Either they accept that there are
substitutes for new steel drums or, alternatively, that their tendering
process would maintain the requisite degree of competition between the
two remaining producers of steel drums. Given the high cost of
transporting empty drums the threat of international procurement does not
seem significant.
23. The manufacture of steel drums is an established technology and –
Sirco’s drum kits notwithstanding – there is no evidence suggesting that
this is a dynamic or innovative market. However, there is, to be sure,
7
evidence of innovation in alternative materials (plastics, most notably).
There is evidence to suggest that producers active in plastic packaging
are increasing their capacity. Given the uncertainty surrounding the
definition of the relevant market, we are unable to reach a firm conclusion
regarding the dynamism of the market – if steel drums are not
substitutable then the technology is mature and the market is relatively
‘undynamic’; if other industrial products are substitutable for new steel
drums then the market appears relatively dynamic and innovative.
24. The parties have made much of the failure of Drumpak. They point out
that prohibition of the merger would not restore Drumpak and that,
accordingly, the level of competition in the market would not be improved
by prohibiting the merger. This argument is rejected. That the operational
aspects of the transaction were completed precipitously by Drumpak
cannot be advanced as adequate ground for the Tribunal to condone an
anticompetitive acquisition. Furthermore, the Tribunal does not question
Drumpak’s right to exit the market or to sell its assets. At issue is the sale
of assets to a consortium of its competitors. Had the Tribunal been
persuaded that these were its only competitors, that is, that the relevant
market was steel drums rather than industrial containers, the fact that
Drumpak had caused itself to fail by effective implementation of the
merger would not have deterred the Tribunal from prohibiting the
transaction.
25. All the evidence confirms that Drumpak has been an effective competitor
in the market. It is common cause that Drumpak’s entry into the market
coincided with a considerable softening of prices for new steel drums. It
coincided with a considerable softening of prices for new steel drums. It
has been suggested that the softening of prices may have occurred in
consequence of the availability of competing products. It has also been
suggested that its ability to discount its price was dependent upon a state
subsidy to which it no longer enjoyed access. Alternatively it may be
argued that Drumpak’s recent troubles were, in common with many other
small producers, a product of general economic malaise. Again, here, the
evidence presented by both parties leaves much to be desired.
26. In summary, the Tribunal has been presented with a welter of inconclusive
evidence. We stress that while transactions of this sort may have
extremely harmful effects on competition and hence are subject to
regulation, merger regulation is not intended to establish culpability on the
part of any party; there is, in other words, no presumption in the Act that
mergers are, per se , likely to inhibit competition and that they are to be
discouraged, much less prevented. On the contrary they may be part of
the necessary process of economic restructuring. Accordingly, when the
8
Commission recommends that a transaction be prohibited, the Tribunal
will insist on a higher standard of evidence than that presented here.
While the parties have failed to establish that their merger will not harm
competition, this is not sufficient to prohibit the transaction.
Public Interest
27. Drumpak employed some 32 workers, most of whom had secured
alternative employment when Drumpak ceased operations. Employment
loss is accordingly minimal. No other public interest issues are relevant in
this case.
28. Accordingly, the Tribunal orders that the transaction be allowed to proceed
unconditionally.
___________________________ ________________
D.H. Lewis Date
Presiding Member
Concurring: N.M. Manoim and P.E. Maponya
9