---

COMPETITION TRIBUNAL

REPUBLIC OF SOUTH AFRICA

Case no.: 17/LM/Mar04

In the large merger between:

Afgri Operations Ltd

and

Natal Agricultural Co-Operative Ltd

______________________________________________________________

Reasons

______________________________________________________________

Introduction

On 4 June 2004 the Competition Tribunal approved the merger between Afgri Operations Ltd (“Afgri”) and Natal Agricultural Co-Operative Ltd (“Natalagri”). The reasons are set out below.

Description of merger

Afgri is acquiring all the shares in the issued share capital of Natalagri and all the members’ funds, other than shares held by and the funds attributable to Afgri and Laeveld, by way of a scheme of arrangement. On implementation of the merger Afgri will control Natalagri.

Rationale

Natalagri has in recent years been experiencing cash flow problems in part due to an unsuccessful acquisition. Placed under pressure by its funders, the co-operative's management recommended to members that they sell the business. Invitations to bidders were submitted and three offers were made. The offer from Afgri was the one accepted by the board.

For its part Afgri has been keen to maker further acquisitions both to expand its geographic footprint and to increase its customer base. Natalagri also owns assets that Afgri is keen to control including its silos, which are well located for the export market.¹

Products and services supplied by the merging parties

The Commission identified the following markets in which both parties provide products and services:

• Financial services products These involve the provision of credit facilities and the brokering of crop and life insurance. The market is considered to be national.

The merged entity’s market share in the provision of credit facilities will be 6.8%. Its main rivals are Absa Bank, with 18% of the market, and the Landbank with 26%. With regard to insurance products there are large firms such as Absa Insurance brokers, FNB Insurance brokers and Senwes who compete with the merged entity. The Commission believes that the transaction is thus unlikely to substantially prevent or lessen competition in the national market.

• Trading of agricultural commodities Agricultural commodities such as grain and maize are traded on Safex (South African Futures Exchange) and according to the Commission the market is therefore national.

The combined market share of the merged entity is 18.2%. The competitors are Seaboard with a market share of 41.8%, Cargill with 20% and Senwes 14%. The Commission believes that the transaction will thus not lead to a substantial lessening of competition in the market. We comment on this more fully below.

• Handling and storage facilities This product market refers to the storing of grain or maize in silo complexes prior to delivery to millers. According to the Commission the geographic markets for silos are local and farmers generally travel no more than 40km to the nearest silo to store their crop. According to the Commission’s analysis there are no product overlaps in the relevant geographic markets. Again we comment on this more fully below.

• Manufacture and distribution of animal feeds Animal feeds are sold in bulk and bagged forms. The Commission submits that bulk sales have an average geographic reach of approximately 150km from the manufacturing plant. Natalagri supplies animal feed mainly in the northern and central KwaZulu Natal areas and Afgri does not compete in these.

• Operating retail outlets Both parties sell a wide range of products including fuels and lubricants, hardware, tyres and batteries, veterinary medicines, spare parts, etc. The market is local and both parties have outlets in Bethlehem, Ermelo, Kroonstad, Senekal.

Although no market share data is available the Commission found that there is competition from other co-operatives such as Senwes, VKB, Boeredienste, BKB and TWK in the particular geographic markets. Moreover barriers to entry are low as minimal capital expenditure and expertise are required. For this reason the Commission considers that the transaction would not lessen competition in these areas.

• Marketing of farming equipment This includes equipment such as tractors, combine harvesters, balers, planters etc. The merging parties have concluded certain exclusive agreements for certain geographic locations. However farmers are not confined to buying equipment in a particular region.

The Commission found that if the market was regional then the merging parties did not compete and that if it was national their market shares would be less than 7%, hence raising no concerns.

• Sale and distribution of crop care products Crop care products are fertilizers and chemicals and are sold in a national market.

The Commission found that not only are there various competitors in these markets, but also that farmers can buy directly from the manufacturers and hence this market raises no competition concerns.

The Commission also investigated the effect of the merger on vertical integration in the seeds market. Afgri operates both a seed manufacturing division, and a financing and a retail division, and is thus vertically integrated in the seeds market. By acquiring Natalagri, which is involved in the financing and retailing of seeds, Agfri will extend its vertical integration into KwaZulu-Natal. The Commission identified two areas where foreclosure could arise, namely 1) Afgri could refuse to finance seeds sales made by rival seed companies, 2) Afgri may foreclose rival seed companies from selling its seeds in Natalagri retail outlets.

With regard to the first concern the Commission found that alternative finance options are available through various banks as well as seed companies that are prepared to finance their own sales to farmers. Regarding the second concern, the Commission found that there is no incentive for Afgri to refuse to stock rivals’ seeds as it would compromise the levels of service in Afgri’s outlets if they did not provide a full range of product to the farmer. Moreover, should foreclosure take place, seed manufacturers do have alternative channels such as hardware stores to sell their seeds. The Commission therefore found that from a vertical perspective the merger did not raise any competition concerns.

Analysis

We have seen this merger quite differently to the Commission. The Commission was most concerned with examining the possibility of vertical foreclosure in the seed markets. Ultimately, and we have no quibble with this, it concluded that the likelihood of foreclosure in this market was remote. We also have no criticism with its analysis of the remaining product services where there were overlaps and its conclusion that none of these raised concerns as any increment in market share when viewed nationally was de minimus and if viewed regionally was unlikely to give rise to overlaps at all. We do not for this reason need to determine the precise boundaries of the geographic markets.

Whether the Commission and the merging parties are right to regard the markets as separate and discrete issues for individual analysis may be open to some doubt. It may be better to view firms such as Afgri and Natalagri as providers of a package of services to a client base. They compete with one another for this client base Just as we approach competition between supermarkets on the basis of competitors selling a package of products to consumers, rather than breaking down a competition analysis in respect of each item in the package, so it might be better to consider this type of merger in future in the same way i.e. a merger between firms that sell a package of products and services to a customer base. This approach also seems to accord with how the parties view themselves and their competitors.²

For instance, in the report prepared for the Afgri acquisition committee, motivating the merger, it is noted that:

“Three Co-operatives (TWK, VKB, and CFC) are operational and active in the existing Natalagri area.

The expectation is that, as is the case in other parts of the country, these Co-operatives will continue to attempt to build their customer base in the KwaZulu Natal region.” ³

That Afgri regards the expansion of its customer base as a strategic issue is not only clear from this extract, but also the fact that it operates a loyalty scheme with other firms known as the Agri Bonus scheme. Customers who purchase from firms who are part of the scheme receive credits that can be redeemed in the form of bonus payments back to them.⁴ This suggests that firms like Afgri compete for the retention of a repeat customer base by offering loyalty incentives to retain customers and disincentivises them from going to rivals. By concentrating only on the services by unbundling them as the Commission has done there is a danger of identifying more substitutes than in reality exist for a customer, who may prefer, partly owing to convenience and partly through the operation of loyalty incentive schemes, to buy a package from one supplier.

What received little emphasis in the Commission’s report was the acquisition of Natalagri’s silos. Natalagri owns approximately 360 000 tons of silo capacity in KZN, at eight different sites. These silos represent 85% of the silo capacity available commercially in that province. Insight into the due diligence indicates that the silos were the target firm’s most valuable assets, not only because their replacement value was enormous, but also because they enjoy a local monopoly.

Flowing from these observations in our view the merger should have been analysed in relation to the following concerns:

1. Could it lead to exclusion of rivals of Afgri by tying up an additional customer base through the use of loyalty schemes?

2. Will the merger lead to an increase in the prices for grain storage and handling?

3. Will the merger, by virtue of the increase in concentration in the grain handling and storage market, lead to the merged firm gaining market power in the related grain trading market or over grain prices?

We will now deal with these concerns separately.

The market for co-operative services

It seems from the evidence that there is no exclusivity between Afgri and its customers who would be free to procure from a rival firm. Although the existence of the loyalty scheme usually disincentivises customers from shopping around it is unlikely that the addition of Natalagri’s 1000 customer base is likely to have this effect.

Market for the handling and storage of grain

Like its acquirer, Natalagri is a product of a highly regulated agricultural sector where co-operatives were organised as local monopolies bolstered by elaborate and complex marketing schemes, which were designed to protect producers, and those down the production chain from the inconvenience of competition. The post 1994 de-regulation of this sector, which we have alluded to in some of our earlier decisions (See The Competition Commission v Patensie Sitrus Beherend, Tribunal Case No 37/CR/Jun01 and SA Fruit Terminals (Pty) Ltd v Portnet and others, Tribunal Case No: 52/IR/Sep01) sought to introduce competition into the sector. Yet as the situation with silos illustrates, eradicating the regulations and arrangements is one thing, undoing geography is another. Short of new entry into the silo markets commercial silos are likely to continue as regional monopolies. The investment required to erect new silos does not justify this, and as its silos appear to be operating below full capacity, nor does the demand.

The merging parties allege that there is a form of substitution in that farmers often erect their own silos.⁵ Whilst this may be correct it is unlikely to be the preferred form of substitution for many farmers but this is not an issue that we have to decide.

Both Afgri and Natalagri offer commercial silo services or what the parties referred to as the market for the handling and storage of grain. Afgri’s silos have a capacity for storage of 3 600 727 tonnes, making it the second largest in the market after Senwes, who have a capacity of 4 585 794 tonnes. Natalagri has a capacity of 359 154 tonnes, putting it sixth on the list in respect of grain storage capacity. But the party’s argue that silo markets are local. Due to high transport costs there is a limit to how far farmers are willing to travel to deliver their grain before the transport costs become prohibitive. Although between the parties and the Commission’s informants this figure was not consistent, it varied between 40 and 60 kilometres.

On that basis no Afgri silo was, argued the merging parties and the Commission, sufficiently close to one of Natalagri’s for them to be considered by any customers as an alternative. We asked the parties to provide us with details of the nearest silo to all those of Natalagri including who owned the silos and the distance that they were apart

From this information it emerges that Afgri has the nearest silo to two of the eight Natalagri silos; Natalagri’s silo at Bergville is 140 km from Afgri’s Harrismith silo and Natalagri’s Winterton silo is 170 km from Afgri’s Harrismith Silo.⁶ However the Afgri silo is sufficiently far away for it not to be regarded as a substitute for a competitively significant number of farmers.

Thus, on the face of it, the parties’ contention, which the Commission accepted, that the geographic markets were separate and thus the merging firms grain storage and handling activities did not overlap, appears correct. However, further information supplied prior to and during the hearing presents a more complex picture.

The parties provided us with a price list for the past three years of the tariffs charged for storage by five of the firms that provide commercial silo facilities, including the merging firms.⁷

This table demonstrates various features:

1. all the firms charge differentiated tariffs depending on the nature of the product being stored. (i.e. maize, soya bean or wheat)

2. firms appear to charge a national rather than a regional tariff

3. firms provide tariffs in relation to the time period over which the grain is stored. Typically they provide a yearly, daily and some a monthly tariff. For daily storage a separate handling tariff is charged although this handling tariff is built into the annual tariff in respect of Afgri and therefore not charged for separately.

4. The tariffs for the three largest firms appear remarkably similar while Natalagri’s tariff appears, in most respects, to be lower than any of the others.

5. Afgri itself in the figures provided seems to charge the highest annual tariff.

6. According Mr De Lange most grain is stored in accordance with the annual tariff and the average period of storage is 4.6 months. When grain is stored for less than a year the annual fee is pro-rated.

Was there anything in the similarity of tariffs and fee structures we asked? It then emerged that there was. We were advised that Safex⁸ provides a recommended tariff. This tariff is calculated annually after negotiations that take place between the millers and the silo owners in a committee known as the Grain Industry Committee. The price arrived at by this committee then becomes the Safex tariff. The Safex tariff while non-binding in nature does appear to influence the manner in which firms price their handling tariffs and hence the degree of similarity contained in the price lists is now explicable

Now, as we noted before, Natalagri prices below the other commercial silo operators and this feature was confirmed in evidence during the hearing.⁹ We asked the merging parties if there was any reason why Natalagri under Afgri control would not raise prices at least up to the Safex tariff. We were advised that this would not happen for two reason one legal and one commercial

The legal reason is that Afgri in terms of the sale agreement had undertaken that it would not during the first year of operation raise prices above CPI. This legal undertaking has since been strengthened as a result of consultation that took place between Afgri and members of Natalagri. The latter were concerned about the short duration of the undertaking and Afgri have agreed to make the undertaking in respect of price increases indefinite.

The commercial reason is that the economics of silo tariffs works differently in KZN than it does in other local markets and this ensures that pricing remains more competitive than the Safex tariff. Farms in KZN are on the whole smaller than in other parts of the country because the land is hilly. Having a smaller crop means that for a smaller amount of expenditure a farmer can erect silo capacity on the farm as a substitute if storage prices increase. Secondly farmers are located closer to the millers who are their customers and hence have less need to use third party storage facilities. Thirdly KZN consumes more grain than it produces which means it is a net importer and this apparently means less grain lies waiting in the province’s silos. These three factors have meant that Natalagri has not optimally utilised its silo capacity, and hence this has kept prices down. We are advised that the merger will not lead to a change in these factors.

Given these assurances - in particular the undertakings that Afgri has given to the members - we are satisfied that Afgri will be constrained from raising storage tariffs beyond CPI in the near future and it is therefore not necessary to consider a condition in this respect.

Effect on markets related to the market for the handling and storage of grain

We must now consider whether Afgri’s increased ownership of silos will result in anti-competitive effects in related markets.

Nationally the ownership of silos is highly concentrated. The share of the top three firms, Senwes, Afgri and Noordwes is estimated to range between 70 – 72%. Pre-merger, Afgri has 23% of the market and post merger it will acquire a further 2,3%. From this perspective the increment is de minimus. Nevertheless one cannot ignore concerns that have been expressed by some academic writers that the increased concentration of silo ownership may have a bearing on prices in related markets. In the related market for grain trading, four major players, one of which is Afgri, dominate the market.

Two economists who have written recently on these markets have expressed the view that there is a relationship between concentration in the silo market and wheat prices. In the one paper prepared for the Competition Commission during its investigation into food prices, Professor Herman van Schalkwyk from the University of the Free State quotes the National Agricultural Marketing Council, which believes that:

“the current level of geographic concentration is unhealthy and given the pivotal position of the bulk silo infrastructure in the deregulated markets for maize and wheat it would be preferable for government to be pro-active rather than reactive in the situation”.“¹⁰

Professor Van Schalkwyk believes that operators could capitalise on the geographical concentration and act in a way that could restrict competition in the market. ¹¹

The second paper written by Neo Chabane of the University of Witwatersrand, expresses the concern that:

“A recent trend in the agricultural sector has been increasing economies of scale. At the same time, the ownership of silos has become more concentrated. The buying up of small farms and silos has lead to a situation where oligopoly conditions exist in the maize market. These conditions may enable collusion.”¹²

The parties’ response at the hearing was to de-link the silo market from trading. The parties suggested that what goes on in the trading market is unrelated to who owns the means of storing grain. Once grain is stored in a silo the owner is issued with a certificate in respect of that quantity. These certificates, like share certificates, are tradable commodities. According to Mr Smith, a single certificate may be traded over 10 times before it is redeemed. Once the buyer collects the grain the certificate is redeemed.

This system, the parties argue, separates the market for grain prices from that for the ownership of silos.

Chabane who acknowledges this separation nevertheless remains concerned:

“It is important to recognise, however, that the firms operating the silos do not necessarily own the grain which is stored. Owners, whether farmers or traders, pay fees for grain storage. But, the silos do have a very important role as market makers, posting prices for the purchase of grain. This function and the way in which information is shared amongst them requires further investigation. As already noted, it is also the vertical integration and the combination of related activities which makes the silos so pivotal in the market.”¹³

This is not an issue we have to decide in this case, given the de minimus increment in concentration brought about by the merger. But it does illustrate that these are complex interrelationships, which should have been further investigated by the Commission, given that these are agricultural markets that may affect the price of foodstuffs purchased by the most vulnerable consumers. This is not to suggest that the outcome would have been different if the investigation had been more thorough, but it would have been more comforting to be sure about that fact than to speculate upon it.

Public interest

The merger will not lead to a loss of employment nor does it implicate any of the other public interest concerns

Conclusion

We find then that there is no evidence on the present record to suggest that the merger will lead to a substantial prevention or lessening of competition. The public interest does not lead to any other conclusion and therefore we approve the merger without conditions.

____________ 06 July 2004

N. Manoim Date

Concurring: D. Lewis, U Bhoola

For Afgri: Johan Brink and Alisen de Villiers from Brink Co

and Le Roux.

For Natal Agri: Johan Smith and Mark Stockhile

For the Commission: Maarten van Hoven and Seema Nunkoo

---